[MR V2] Migrate to Model Runner V2

Squashed migration rebased onto current main. Collapses the 46-commit
dev/migrate-MR-v2 history (10 merge commits + superseded pre-rebase work)
into a single commit, so the diff reflects only the genuine MR-V2 feature
(worker_v2/ runners + model_states, omni schedulers, stage engine pipeline,
pooling_output serialization bridge). Content already merged to main
separately (step_audio2, tts_adapters, npu/_310p, etc.) drops out.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

.buildkite/bootstrap-amd-omni.sh

@@ -6,8 +6,6 @@
 
 set -euo pipefail
 
-source .buildkite/common/scripts/resolve_skip_ci.sh
-
 if [[ -z "${RUN_ALL:-}" ]]; then
     RUN_ALL=0
 fi
@@ -24,6 +22,10 @@ if [[ -z "${AMD_MIRROR_HW:-}" ]]; then
     AMD_MIRROR_HW="amdproduction"
 fi
 
+if [[ -z "${DOCS_ONLY_DISABLE:-}" ]]; then
+    DOCS_ONLY_DISABLE=0
+fi
+
 fail_fast() {
     DISABLE_LABEL="ci-no-fail-fast"
     # If BUILDKITE_PULL_REQUEST != "false", then we check the PR labels using curl and jq
@@ -54,6 +56,88 @@ check_run_all_label() {
     fi
 }
 
+is_docs_only_change() {
+    local file_path
+    local has_any=0
+
+    while IFS= read -r file_path; do
+        [[ -z "${file_path}" ]] && continue
+        has_any=1
+
+        if [[ "${file_path}" == docs/* ]]; then
+            continue
+        fi
+        if [[ "${file_path}" == *.md ]]; then
+            continue
+        fi
+        if [[ "${file_path}" == "mkdocs.yml" ]]; then
+            continue
+        fi
+        return 1
+    done
+
+    [[ "${has_any}" -eq 1 ]]
+}
+
+resolve_skip_ci() {
+    local is_pr_build=0
+    local files
+    local base_branch base_ref
+
+    if [[ "${BUILDKITE_PULL_REQUEST:-false}" != "false" && -n "${BUILDKITE_PULL_REQUEST:-}" ]]; then
+        is_pr_build=1
+    fi
+
+    if [[ "${is_pr_build}" -eq 1 ]]; then
+        base_branch="${BUILDKITE_PULL_REQUEST_BASE_BRANCH:-main}"
+        if ! git rev-parse --verify "origin/${base_branch}" >/dev/null 2>&1; then
+            echo "resolve_skip_ci: origin/${base_branch} not found locally; trying fetch" >&2
+            git fetch --depth=200 origin "${base_branch}" >/dev/null 2>&1 || true
+        fi
+
+        base_ref=""
+        if git rev-parse --verify "origin/${base_branch}" >/dev/null 2>&1; then
+            base_ref="origin/${base_branch}"
+        elif git rev-parse --verify "${base_branch}" >/dev/null 2>&1; then
+            base_ref="${base_branch}"
+        else
+            echo "resolve_skip_ci: cannot resolve PR base ${base_branch}; skip-ci=0" >&2
+            echo -n 0
+            return 0
+        fi
+
+        if ! files="$(git diff --name-only "${base_ref}...${BUILDKITE_COMMIT}" 2>/dev/null)"; then
+            echo "resolve_skip_ci: failed to compute PR changed files; skip-ci=0" >&2
+            echo -n 0
+            return 0
+        fi
+    elif [[ "${BUILDKITE_BRANCH:-}" == "main" ]]; then
+        if ! git rev-parse --verify "${BUILDKITE_COMMIT}^" >/dev/null 2>&1; then
+            echo "resolve_skip_ci: commit has no parent on main; skip-ci=0" >&2
+            echo -n 0
+            return 0
+        fi
+        if ! files="$(git diff --name-only "${BUILDKITE_COMMIT}^..${BUILDKITE_COMMIT}" 2>/dev/null)"; then
+            echo "resolve_skip_ci: failed to compute main changed files; skip-ci=0" >&2
+            echo -n 0
+            return 0
+        fi
+    else
+        echo "resolve_skip_ci: not PR/main build; skip-ci=0" >&2
+        echo -n 0
+        return 0
+    fi
+
+    if is_docs_only_change <<< "${files}"; then
+        echo "resolve_skip_ci: docs-only change detected; skip-ci=1" >&2
+        echo -n 1
+        return 0
+    fi
+
+    echo "resolve_skip_ci: non-doc changes detected; skip-ci=0" >&2
+    echo -n 0
+}
+
 if [[ -z "${COV_ENABLED:-}" ]]; then
     COV_ENABLED=0
 fi
@@ -70,14 +154,14 @@ upload_pipeline() {
     fi
 
     # Use local template file for vllm-omni
-    cp .buildkite/amd/test-template-amd-omni.j2 .buildkite/amd/test-template.j2
+    cp .buildkite/test-template-amd-omni.j2 .buildkite/test-template.j2
 
 
     # (WIP) Use pipeline generator instead of jinja template
-    if [ -e ".buildkite/amd/pipeline_generator/pipeline_generator.py" ]; then
+    if [ -e ".buildkite/pipeline_generator/pipeline_generator.py" ]; then
         python -m pip install click pydantic
-        python .buildkite/amd/pipeline_generator/pipeline_generator.py --run_all=$RUN_ALL --list_file_diff="$LIST_FILE_DIFF" --nightly="$NIGHTLY" --mirror_hw="$AMD_MIRROR_HW"
-        buildkite-agent pipeline upload .buildkite/amd/pipeline.yaml
+        python .buildkite/pipeline_generator/pipeline_generator.py --run_all=$RUN_ALL --list_file_diff="$LIST_FILE_DIFF" --nightly="$NIGHTLY" --mirror_hw="$AMD_MIRROR_HW"
+        buildkite-agent pipeline upload .buildkite/pipeline.yaml
         exit 0
     fi
     echo "List file diff: $LIST_FILE_DIFF"
@@ -87,7 +171,7 @@ upload_pipeline() {
 
     FAIL_FAST=$(fail_fast)
 
-    cd .buildkite/amd
+    cd .buildkite
 
     # Select test definition file: merge suite for main, ready suite for PRs.
     # For debugging, DEBUG_TEST_YAML can override the selection — accepts
@@ -98,7 +182,7 @@ upload_pipeline() {
                 TEST_YAML="test-amd-merge.yml"
                 ;;
             ready)
-                TEST_YAML="test-amd-ready.yml"
+                TEST_YAML="test-amd-ready.yaml"
                 ;;
             *)
                 echo "ERROR: DEBUG_TEST_YAML must be 'merge' or 'ready', got '$DEBUG_TEST_YAML'" >&2
@@ -109,7 +193,7 @@ upload_pipeline() {
     elif [[ $BUILDKITE_BRANCH == "main" ]]; then
         TEST_YAML="test-amd-merge.yml"
     else
-        TEST_YAML="test-amd-ready.yml"
+        TEST_YAML="test-amd-ready.yaml"
     fi
 
     (
@@ -150,13 +234,23 @@ if [[ $BUILDKITE_BRANCH == "main" ]]; then
     file_diff=$(get_diff_main)
 fi
 
-# Early exit: unified skip-ci (docs / skip marks) and CI-yaml-only level targeting.
-if [[ $BUILDKITE_BRANCH == "main" ]]; then
-    gate_bootstrap_ci amd l3
-else
-    gate_bootstrap_ci amd l2
+# ----------------------------------------------------------------------
+# Early exit start: skip pipeline if conditions are met
+# ----------------------------------------------------------------------
+
+# Match CUDA Buildkite skip-ci behavior for docs/markdown-only changes.
+if [[ "${DOCS_ONLY_DISABLE}" != "1" ]]; then
+    SKIP_CI="$(resolve_skip_ci)"
+    if [[ "${SKIP_CI}" == "1" ]]; then
+        echo "[docs-only] Docs/markdown-only changes detected. Exiting before pipeline upload."
+        exit 0
+    fi
 fi
 
+# ----------------------------------------------------------------------
+# Early exit end
+# ----------------------------------------------------------------------
+
 patterns=(
     "docker/Dockerfile"
     "CMakeLists.txt"

.buildkite/bootstrap-intel-omni.sh

@@ -0,0 +1,51 @@
+#!/bin/bash
+# vllm-omni Intel bootstrap
+# Uses static pipeline-intel.yaml for Intel XPU tests
+# Based on: bootstrap-amd-omni.sh
+
+set -euo pipefail
+
+if [[ -z "${DOCS_ONLY_DISABLE:-}" ]]; then
+    DOCS_ONLY_DISABLE=0
+fi
+
+upload_pipeline() {
+    echo "--- 🛠 Preparing Intel pipeline"
+    buildkite-agent pipeline upload .buildkite/pipeline-intel.yaml
+}
+
+get_diff() {
+    $(git add .)
+    echo $(git diff --name-only --diff-filter=ACMDR $(git merge-base origin/main HEAD))
+}
+
+get_diff_main() {
+    $(git add .)
+    echo $(git diff --name-only --diff-filter=ACMDR HEAD~1)
+}
+
+file_diff=$(get_diff)
+if [[ $BUILDKITE_BRANCH == "main" ]]; then
+    file_diff=$(get_diff_main)
+fi
+
+# ----------------------------------------------------------------------
+# Early exit: skip pipeline if all changed files are under docs/
+# ----------------------------------------------------------------------
+if [[ "${DOCS_ONLY_DISABLE}" != "1" ]] && [[ -n "${file_diff:-}" ]]; then
+    docs_only=1
+    while IFS= read -r f; do
+        [[ -z "$f" ]] && continue
+        if [[ "$f" != docs/* ]]; then
+            docs_only=0
+            break
+        fi
+    done < <(printf '%s\n' "$file_diff" | tr ' ' '\n' | tr -d '\r')
+
+    if [[ "$docs_only" -eq 1 ]]; then
+        buildkite-agent annotate ":memo: CI skipped — docs only" --style "info"
+        exit 0
+    fi
+fi
+
+upload_pipeline

.buildkite/common/ci_mirror_hardwares.yml

@@ -1,239 +0,0 @@
-# Uploader-only hardware presets for CUDA / NPU test YAML.
-# Referenced as ``mirror_hardwares: l4_1`` on a step; ``upload_pipeline.py`` expands to
-# ``agents`` (+ optional top-level ``image`` for NPU) + ``plugins``, then strips
-# ``mirror_hardwares`` before Buildkite upload.
-#
-# CUDA naming: {chip}_{count}
-#   l4_1, l4_4, h100_1 … h100_4
-# NPU naming: {queue}_{resource_class}
-#   a2b3_npu_1  — ascend-a2b3, resource_class npu-1 (A2/B3 CI image)
-#   a2b3_npu_4  — ascend-a2b3, resource_class npu-4
-#   a3_npu_2    — ascend-a3, resource_class npu-2 (A3 CI image tag …:$BUILDKITE_COMMIT-a3)
-
-x-npu-k8s-plugins: &npu_k8s_plugins
-  - kubernetes:
-      podSpecPatch:
-        imagePullSecrets:
-          - name: swr-secret
-
-
-mirror_hardwares:
-  l4_1:
-    agents:
-      queue: gpu_1_queue
-    plugins:
-      - docker#v5.2.0:
-          image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-          always-pull: true
-          propagate-environment: true
-          shm-size: "8gb"
-          environment:
-            - HF_HOME=/fsx/hf_cache
-            - HF_TOKEN
-          volumes:
-            - /fsx/hf_cache:/fsx/hf_cache
-
-  l4_4:
-    agents:
-      queue: gpu_4_queue
-    plugins:
-      - docker#v5.2.0:
-          image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-          always-pull: true
-          propagate-environment: true
-          shm-size: "8gb"
-          environment:
-            - HF_HOME=/fsx/hf_cache
-            - HF_TOKEN
-          volumes:
-            - /fsx/hf_cache:/fsx/hf_cache
-
-  h100_1:
-    agents:
-      queue: mithril-h100-pool
-    plugins:
-      - kubernetes:
-          podSpec:
-            containers:
-              - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-                resources:
-                  limits:
-                    nvidia.com/gpu: 1
-                volumeMounts:
-                  - name: devshm
-                    mountPath: /dev/shm
-                  - name: hf-cache
-                    mountPath: /root/.cache/huggingface
-                env:
-                  - name: HF_HOME
-                    value: /root/.cache/huggingface
-                  - name: HF_TOKEN
-                    valueFrom:
-                      secretKeyRef:
-                        name: hf-token-secret
-                        key: token
-            nodeSelector:
-              node.kubernetes.io/instance-type: gpu-h100-sxm
-            volumes:
-              - name: devshm
-                emptyDir:
-                  medium: Memory
-              - name: hf-cache
-                hostPath:
-                  path: /mnt/hf-cache
-                  type: DirectoryOrCreate
-
-  h100_2:
-    agents:
-      queue: mithril-h100-pool
-    plugins:
-      - kubernetes:
-          podSpec:
-            containers:
-              - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-                resources:
-                  limits:
-                    nvidia.com/gpu: 2
-                volumeMounts:
-                  - name: devshm
-                    mountPath: /dev/shm
-                  - name: hf-cache
-                    mountPath: /root/.cache/huggingface
-                env:
-                  - name: HF_HOME
-                    value: /root/.cache/huggingface
-                  - name: HF_TOKEN
-                    valueFrom:
-                      secretKeyRef:
-                        name: hf-token-secret
-                        key: token
-            nodeSelector:
-              node.kubernetes.io/instance-type: gpu-h100-sxm
-            volumes:
-              - name: devshm
-                emptyDir:
-                  medium: Memory
-              - name: hf-cache
-                hostPath:
-                  path: /mnt/hf-cache
-                  type: DirectoryOrCreate
-
-  h100_3:
-    agents:
-      queue: mithril-h100-pool
-    plugins:
-      - kubernetes:
-          podSpec:
-            containers:
-              - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-                resources:
-                  limits:
-                    nvidia.com/gpu: 3
-                volumeMounts:
-                  - name: devshm
-                    mountPath: /dev/shm
-                  - name: hf-cache
-                    mountPath: /root/.cache/huggingface
-                env:
-                  - name: HF_HOME
-                    value: /root/.cache/huggingface
-                  - name: HF_TOKEN
-                    valueFrom:
-                      secretKeyRef:
-                        name: hf-token-secret
-                        key: token
-            nodeSelector:
-              node.kubernetes.io/instance-type: gpu-h100-sxm
-            volumes:
-              - name: devshm
-                emptyDir:
-                  medium: Memory
-              - name: hf-cache
-                hostPath:
-                  path: /mnt/hf-cache
-                  type: DirectoryOrCreate
-
-  h100_4:
-    agents:
-      queue: mithril-h100-pool
-    plugins:
-      - kubernetes:
-          podSpec:
-            containers:
-              - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-                resources:
-                  limits:
-                    nvidia.com/gpu: 4
-                volumeMounts:
-                  - name: devshm
-                    mountPath: /dev/shm
-                  - name: hf-cache
-                    mountPath: /root/.cache/huggingface
-                env:
-                  - name: HF_HOME
-                    value: /root/.cache/huggingface
-                  - name: HF_TOKEN
-                    valueFrom:
-                      secretKeyRef:
-                        name: hf-token-secret
-                        key: token
-            nodeSelector:
-              node.kubernetes.io/instance-type: gpu-h100-sxm
-            volumes:
-              - name: devshm
-                emptyDir:
-                  medium: Memory
-              - name: hf-cache
-                hostPath:
-                  path: /mnt/hf-cache
-                  type: DirectoryOrCreate
-
-  a2b3_npu_1:
-    agents:
-      queue: "ascend-a2b3"
-      resource_class: "npu-1"
-    image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}"
-    plugins: *npu_k8s_plugins
-
-  a2b3_npu_4:
-    agents:
-      queue: "ascend-a2b3"
-      resource_class: "npu-4"
-    image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}"
-    plugins: *npu_k8s_plugins
-
-
-  a2b3_npu_8:
-    agents:
-      queue: "ascend-a2b3"
-      resource_class: "npu-8"
-    image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}"
-    plugins: *npu_k8s_plugins
-
-  a3_npu_2:
-    agents:
-      queue: "ascend-a3"
-      resource_class: "npu-2"
-    image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}-a3"
-    plugins: *npu_k8s_plugins
-
-  a3_npu_4:
-    agents:
-      queue: "ascend-a3"
-      resource_class: "npu-4"
-    image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}-a3"
-    plugins: *npu_k8s_plugins
-
-  a3_npu_8:
-    agents:
-      queue: "ascend-a3"
-      resource_class: "npu-8"
-    image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}-a3"
-    plugins: *npu_k8s_plugins
-
-  a3_npu_16:
-    agents:
-      queue: "ascend-a3"
-      resource_class: "npu-16"
-    image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}-a3"
-    plugins: *npu_k8s_plugins

.buildkite/common/scripts/resolve_skip_ci.sh

@@ -1,23 +0,0 @@
-#!/usr/bin/env bash
-# Shared Buildkite skip-ci helpers for AMD / Intel bootstrap scripts.
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-SKIP_CI_PY="${SCRIPT_DIR}/skip_ci.py"
-
-# Resolve git diff once. Exit 0 when this bootstrap should stop uploading.
-# Decision details are logged by skip_ci.py; this helper only annotates skip-all.
-# Usage: gate_bootstrap_ci <platform> <l2|l3>
-gate_bootstrap_ci() {
-    local platform="$1"
-    local level="$2"
-    local kind
-    if kind="$(python3 "${SKIP_CI_PY}" gate "${platform}" "${level}")"; then
-        if [[ "${kind}" == "skip-all" ]]; then
-            buildkite-agent annotate \
-                ":memo: CI skipped — docs or pytest skip-mark changes only" \
-                --style "info" 2>/dev/null || true
-        fi
-        exit 0
-    fi
-}

.buildkite/common/scripts/skip_ci.py

@@ -1,617 +0,0 @@
-#!/usr/bin/env python3
-# SPDX-License-Identifier: Apache-2.0
-"""Shared Buildkite skip-ci and CI-YAML level targeting logic.
-
-``resolve_ci_decision(changed_files, ...)`` buckets explicit file lists (tests, dry-run).
-``resolve_ci_context_from_git()`` resolves git diff once and returns decision + changed files
-for Buildkite bootstrap / upload scripts.
-
-CLI (for bootstrap scripts; exit code is the signal):
-  python3 skip_ci.py gate <platform> <level>
-      Exit 0 when bootstrap should stop (skip-all, or that L2/L3 target is off).
-      Prints ``skip-all`` or ``skip-l23`` on stdout; logs the decision once.
-  python3 skip_ci.py check-skip-all
-      Exit 0 when only docs/skip-mark changes → bootstrap skips entire CI upload.
-  python3 skip_ci.py check-skip-l2-l3 <platform> <level>
-      Exit 0 when that L2/L3 pipeline should be skipped (CI config YAML-only diffs).
-  python3 skip_ci.py print-annotate
-      Print decision message for Buildkite annotation.
-"""
-
-from __future__ import annotations
-
-import argparse
-import os
-import re
-import subprocess
-import sys
-from dataclasses import dataclass, field, replace
-from pathlib import Path
-
-LOG = "[skip-ci]"
-ROOT = Path(__file__).resolve().parent.parent.parent.parent
-
-PLATFORMS = ("cuda", "amd", "intel", "npu")
-L23_LEVELS = frozenset({"l2", "l3"})
-
-L2_YAML_FILES: dict[str, str] = {
-    ".buildkite/cuda/test-ready.yml": "cuda",
-    ".buildkite/npu/test-npu-ready.yml": "npu",
-    ".buildkite/amd/test-amd-ready.yml": "amd",
-    ".buildkite/intel/pipeline-intel.yml": "intel",
-}
-
-L3_YAML_FILES: dict[str, str] = {
-    ".buildkite/cuda/test-merge.yml": "cuda",
-    ".buildkite/amd/test-amd-merge.yml": "amd",
-}
-
-L45_YAML_FILES: dict[str, str] = {
-    ".buildkite/cuda/test-nightly.yml": "cuda",
-    ".buildkite/cuda/test-weekly.yml": "cuda",
-    ".buildkite/npu/test-npu-nightly.yml": "npu",
-}
-
-
-def _empty_platform_buckets() -> dict[str, list[str]]:
-    return {platform: [] for platform in PLATFORMS}
-
-
-@dataclass(frozen=True)
-class CiDevice:
-    """Per-platform / per-level skip flags for yaml-gated bootstrap (``skip_l2_l3``)."""
-
-    skip_cuda_l2: bool = False
-    skip_cuda_l3: bool = False
-    skip_amd_l2: bool = False
-    skip_amd_l3: bool = False
-    skip_intel_l2: bool = False
-    skip_intel_l3: bool = False
-    skip_npu_l2: bool = False
-    skip_npu_l3: bool = False
-
-    def is_skip(self, platform: str, level: str) -> bool:
-        return bool(getattr(self, f"skip_{platform}_{level}", True))
-
-
-@dataclass
-class DiffBuckets:
-    """Changed files grouped by doc / skip-mark / CI yaml / other."""
-
-    docs: list[str] = field(default_factory=list)
-    skip_tests: list[str] = field(default_factory=list)
-    other: list[str] = field(default_factory=list)
-    l2: dict[str, list[str]] = field(default_factory=_empty_platform_buckets)
-    l3: dict[str, list[str]] = field(default_factory=_empty_platform_buckets)
-    l45: dict[str, list[str]] = field(default_factory=_empty_platform_buckets)
-
-    @property
-    def has_doc_changes(self) -> bool:
-        return bool(self.docs)
-
-    @property
-    def has_skip_test_changes(self) -> bool:
-        return bool(self.skip_tests)
-
-    @property
-    def has_other_changes(self) -> bool:
-        """Product code or paths outside skip-ci scope."""
-        return bool(self.other)
-
-    @property
-    def has_l23_yaml_changes(self) -> bool:
-        """At least one whitelisted L2/L3 CI yaml file changed."""
-        return any(self.l2[platform] or self.l3[platform] for platform in PLATFORMS)
-
-    @property
-    def has_l45_yaml_changes(self) -> bool:
-        """At least one whitelisted L4/L5 (nightly/weekly) CI yaml file changed."""
-        return any(self.l45[platform] for platform in PLATFORMS)
-
-    @property
-    def has_skip_ci_scope_changes(self) -> bool:
-        """At least one doc / skip-mark / whitelisted CI yaml change."""
-        return (
-            self.has_doc_changes or self.has_skip_test_changes or self.has_l23_yaml_changes or self.has_l45_yaml_changes
-        )
-
-    def l2_platforms(self) -> frozenset[str]:
-        return frozenset(platform for platform in PLATFORMS if self.l2[platform])
-
-    def l3_platforms(self) -> frozenset[str]:
-        return frozenset(platform for platform in PLATFORMS if self.l3[platform])
-
-
-@dataclass(frozen=True)
-class SkipL2L3Basis:
-    skip_all_l2: bool
-    skip_all_l3: bool
-
-    @classmethod
-    def from_buckets(cls, buckets: DiffBuckets) -> SkipL2L3Basis | None:
-        """L4/L5 yaml → skip both L2 and L3 (L2/L3 buckets may rescue).
-
-        Docs / skip-mark changes do not block this path; they are ignored for
-        L2/L3 targeting when CI YAML is also present.
-        """
-        if buckets.has_other_changes:
-            return None
-        if not buckets.has_l45_yaml_changes:
-            return None
-        return cls(
-            skip_all_l2=not buckets.l2_platforms(),
-            skip_all_l3=not buckets.l3_platforms(),
-        )
-
-
-@dataclass(frozen=True)
-class SkipL3Basis:
-    enable_l2_platforms: frozenset[str]
-
-    @classmethod
-    def from_buckets(cls, buckets: DiffBuckets) -> SkipL3Basis | None:
-        """L2 yaml changed → skip L3; keep L2 on touched platforms."""
-        platforms = buckets.l2_platforms()
-        if not platforms:
-            return None
-        return cls(enable_l2_platforms=platforms)
-
-
-@dataclass(frozen=True)
-class SkipL2Basis:
-    enable_l3_platforms: frozenset[str]
-
-    @classmethod
-    def from_buckets(cls, buckets: DiffBuckets) -> SkipL2Basis | None:
-        """L3 yaml changed → skip L2; keep L3 on touched platforms."""
-        platforms = buckets.l3_platforms()
-        if not platforms:
-            return None
-        return cls(enable_l3_platforms=platforms)
-
-
-@dataclass(frozen=True)
-class CiDecision:
-    """Final CI skip decision for one diff."""
-
-    skip_all: bool = False
-    skip_l2_l3: bool = False
-    device: CiDevice = field(default_factory=CiDevice)
-    message: str = ""
-
-    @classmethod
-    def skip_all_basis(cls, buckets: DiffBuckets, *, diff_range: str | None) -> bool:
-        """Docs / pytest skip-mark only → suppress entire default CI."""
-        if buckets.has_other_changes:
-            return False
-        if (buckets.has_l23_yaml_changes or buckets.has_l45_yaml_changes) and (
-            buckets.has_doc_changes or buckets.has_skip_test_changes
-        ):
-            return False
-        if not buckets.has_doc_changes and not buckets.has_skip_test_changes:
-            return False
-        if diff_range is None and not buckets.has_doc_changes:
-            return False
-        return True
-
-    @classmethod
-    def from_yaml_basis(
-        cls,
-        l2_l3_basis: SkipL2L3Basis | None,
-        l3_basis: SkipL3Basis | None,
-        l2_basis: SkipL2Basis | None,
-    ) -> CiDecision:
-        """Build a yaml-gated decision by merging L2/L3 skip bases.
-
-        Bases are applied as a union of enables: L2 yaml keeps L2 on touched
-        platforms, L3 yaml keeps L3 on touched platforms. Cross-platform mixes
-        (e.g. CUDA ready + AMD merge) must preserve both enables. The resulting
-        per-platform / per-level matrix is stored on ``device`` as ``skip_*``.
-        """
-        run = {platform: {"l2": True, "l3": True} for platform in PLATFORMS}
-
-        if l2_l3_basis is not None:
-            if l2_l3_basis.skip_all_l2:
-                for platform in PLATFORMS:
-                    run[platform]["l2"] = False
-            if l2_l3_basis.skip_all_l3:
-                for platform in PLATFORMS:
-                    run[platform]["l3"] = False
-
-        if l3_basis is not None:
-            for platform in PLATFORMS:
-                run[platform]["l3"] = False
-                run[platform]["l2"] = platform in l3_basis.enable_l2_platforms
-
-        if l2_basis is not None:
-            for platform in PLATFORMS:
-                if platform in l2_basis.enable_l3_platforms:
-                    run[platform]["l3"] = True
-                    if l3_basis is None or platform not in l3_basis.enable_l2_platforms:
-                        run[platform]["l2"] = False
-                elif l3_basis is None:
-                    # L3-yaml-only: disable platforms that did not change L3.
-                    run[platform]["l2"] = False
-                    run[platform]["l3"] = False
-                # else: leave L2 enables from SkipL3Basis; keep this platform's L3 off.
-
-        return cls(
-            skip_l2_l3=True,
-            device=cls._device_from_run(run),
-        )
-
-    @staticmethod
-    def _device_from_run(run: dict[str, dict[str, bool]]) -> CiDevice:
-        return CiDevice(
-            **{f"skip_{platform}_{level}": not run[platform][level] for platform in PLATFORMS for level in ("l2", "l3")}
-        )
-
-    def is_run(self, platform: str, level: str) -> bool:
-        if self.skip_all:
-            return False
-        if not self.skip_l2_l3:
-            return True
-        if level not in L23_LEVELS:
-            return True
-        return not self.device.is_skip(platform, level)
-
-
-@dataclass(frozen=True)
-class CiContext:
-    """Skip-ci decision plus git diff inputs used to compute it."""
-
-    decision: CiDecision
-    changed_files: list[str] | None
-    diff_range: str | None
-
-
-def _log(message: str) -> None:
-    print(f"{LOG}: {message}", file=sys.stderr)
-
-
-def _git(*args: str) -> subprocess.CompletedProcess[str]:
-    return subprocess.run(
-        ["git", *args],
-        check=False,
-        capture_output=True,
-        text=True,
-    )
-
-
-# --- Diff helpers ---
-
-
-def _resolve_diff_range() -> str | None:
-    """Return a git diff range for PR or main builds, or None when unavailable."""
-    is_pr = os.environ.get("BUILDKITE_PULL_REQUEST", "false") != "false" and os.environ.get(
-        "BUILDKITE_PULL_REQUEST", ""
-    )
-    commit = os.environ.get("BUILDKITE_COMMIT", "")
-
-    if is_pr:
-        base_branch = os.environ.get("BUILDKITE_PULL_REQUEST_BASE_BRANCH", "main")
-        base_ref = f"origin/{base_branch}"
-        if _git("rev-parse", "--verify", base_ref).returncode != 0:
-            _log(f"origin/{base_branch} not found locally; trying fetch")
-            _git("fetch", "--depth=200", "origin", base_branch)
-        if _git("rev-parse", "--verify", base_ref).returncode != 0:
-            if _git("rev-parse", "--verify", base_branch).returncode == 0:
-                base_ref = base_branch
-            else:
-                _log(f"cannot resolve PR base {base_branch}; using safe defaults")
-                return None
-        return f"{base_ref}...{commit}"
-
-    if os.environ.get("BUILDKITE_BRANCH", "") == "main":
-        if _git("rev-parse", "--verify", f"{commit}^").returncode != 0:
-            _log("main commit has no parent; using safe defaults")
-            return None
-        return f"{commit}^..{commit}"
-
-    _log("not PR/main build; using safe defaults")
-    return None
-
-
-def _changed_files_for_diff_range(diff_range: str | None) -> list[str] | None:
-    """Return changed file paths for *diff_range*, or None when unavailable."""
-    if diff_range is None:
-        return None
-
-    result = _git("diff", "--name-only", diff_range)
-    if result.returncode != 0:
-        _log(f"git diff failed ({diff_range}); using safe defaults")
-        return None
-
-    files = [line for line in result.stdout.splitlines() if line.strip()]
-    _log(f"{len(files)} changed file(s)")
-    return files
-
-
-def resolve_ci_context_from_git() -> CiContext:
-    """Resolve git diff once and return skip-ci decision plus changed files."""
-    diff_range = _resolve_diff_range()
-    changed_files = _changed_files_for_diff_range(diff_range)
-    decision = resolve_ci_decision(changed_files, diff_range=diff_range)
-    return CiContext(decision=decision, changed_files=changed_files, diff_range=diff_range)
-
-
-# --- Doc / skip-mark helpers ---
-
-
-_SKIP_MARK_RE = re.compile(r"pytest\.mark\.skip(?:if)?\b|pytest\.skip\s*\(")
-_PYTESTMARK_SKIP_RE = re.compile(r"pytest\.mark\.skip\b")
-_PYTEST_MARK_ONLY_RE = re.compile(r"pytest\.mark\.\w+")
-
-
-def _diff_only_contains_skip_mark_changes(diff_text: str) -> bool:
-    """Return True when a unified diff only *adds* pytest skip marks.
-
-    Deleting a skip mark re-enables the test and must not suppress CI. Editing an
-    existing skip (``-`` old / ``+`` new) still qualifies when at least one skip
-    mark is added.
-    """
-
-    def paren_balance(line: str) -> int:
-        return line.count("(") - line.count(")")
-
-    def is_blank_or_comment(line: str) -> bool:
-        stripped = line.strip()
-        return not stripped or stripped.startswith("#")
-
-    def is_skip_mark_line(line: str) -> bool:
-        return _SKIP_MARK_RE.search(line.strip()) is not None
-
-    def is_pytestmark_adjacent_line(line: str) -> bool:
-        stripped = line.strip().rstrip(",")
-        if not stripped or stripped in {"[", "]"} or stripped.startswith("#"):
-            return True
-        if stripped.startswith("pytestmark"):
-            return True
-        return _PYTEST_MARK_ONLY_RE.search(stripped) is not None
-
-    pending_depth = 0
-    saw_change = False
-    has_added_skip_mark = False
-    for raw_line in diff_text.splitlines():
-        if raw_line.startswith("@@"):
-            pending_depth = 0
-            continue
-        if not (raw_line.startswith("+") or raw_line.startswith("-")):
-            continue
-        if raw_line.startswith("+++") or raw_line.startswith("---"):
-            continue
-
-        saw_change = True
-        is_add = raw_line.startswith("+")
-        content = raw_line[1:]
-
-        if pending_depth > 0:
-            pending_depth += paren_balance(content)
-            if pending_depth < 0:
-                pending_depth = 0
-            continue
-
-        if is_blank_or_comment(content):
-            continue
-
-        if is_skip_mark_line(content):
-            if is_add:
-                has_added_skip_mark = True
-            pending_depth = max(0, paren_balance(content))
-            continue
-
-        if is_pytestmark_adjacent_line(content):
-            continue
-
-        return False
-
-    return saw_change and has_added_skip_mark
-
-
-def _is_skip_testcase_change(file_path: str, *, diff_range: str) -> bool:
-    """Return True when a test file diff is skip-mark-only (including new-file cases)."""
-    diff_result = _git("diff", diff_range, "--", file_path)
-    if diff_result.returncode != 0:
-        _log(f"skip-mark-only: git diff failed for {file_path}")
-        return False
-
-    diff_text = diff_result.stdout
-    if not diff_text.strip():
-        _log(f"skip-mark-only: empty diff for {file_path}")
-        return False
-
-    if _diff_only_contains_skip_mark_changes(diff_text):
-        return True
-
-    if "new file mode" not in diff_text and "--- /dev/null" not in diff_text:
-        _log(f"skip-mark-only: non-skip changes in {file_path}")
-        return False
-
-    commit = os.environ.get("BUILDKITE_COMMIT", "")
-    if not commit:
-        return False
-
-    content_result = _git("show", f"{commit}:{file_path}")
-    if content_result.returncode != 0:
-        _log(f"skip-mark-only: cannot read {file_path} at {commit}")
-        return False
-
-    content = content_result.stdout
-    match = re.search(r"^pytestmark\s*=\s*\[(.*?)^\s*\]", content, re.MULTILINE | re.DOTALL)
-    if match is not None and _PYTESTMARK_SKIP_RE.search(match.group(1)) is not None:
-        _log(f"skip-mark-only: new test file with module-level pytest.mark.skip: {file_path}")
-        return True
-    if re.search(r"^pytestmark\s*=\s*pytest\.mark\.skip\b", content, re.MULTILINE):
-        _log(f"skip-mark-only: new test file with module-level pytest.mark.skip: {file_path}")
-        return True
-
-    _log(f"skip-mark-only: new test file without module-level pytest.mark.skip: {file_path}")
-    return False
-
-
-def _classify_changed_files_into_buckets(changed_files: list[str], *, diff_range: str | None) -> DiffBuckets:
-    """Classify each changed file path into a ``DiffBuckets`` category."""
-    buckets = DiffBuckets()
-    for file_path in changed_files:
-        if not file_path:
-            continue
-        if file_path.startswith("docs/") or file_path.endswith(".md") or file_path == "mkdocs.yml":
-            buckets.docs.append(file_path)
-            continue
-        path = Path(file_path)
-        if path.suffix == ".py" and path.parts and path.parts[0] == "tests":
-            if diff_range is not None and _is_skip_testcase_change(file_path, diff_range=diff_range):
-                buckets.skip_tests.append(file_path)
-            else:
-                buckets.other.append(file_path)
-            continue
-        if file_path in L2_YAML_FILES:
-            platform = L2_YAML_FILES[file_path]
-            buckets.l2[platform].append(file_path)
-        elif file_path in L3_YAML_FILES:
-            platform = L3_YAML_FILES[file_path]
-            buckets.l3[platform].append(file_path)
-        elif file_path in L45_YAML_FILES:
-            platform = L45_YAML_FILES[file_path]
-            buckets.l45[platform].append(file_path)
-        else:
-            buckets.other.append(file_path)
-    return buckets
-
-
-# --- Decision ---
-
-
-def _finish(decision: CiDecision, message: str) -> CiDecision:
-    finished = replace(decision, message=message)
-    _log(message)
-    return finished
-
-
-def _format_yaml_gated_message(buckets: DiffBuckets, decision: CiDecision) -> str:
-    """Human-readable summary of which CI YAML changed and which L2/L3 targets run."""
-    changed_parts: list[str] = []
-    for level_name, by_platform in (("L2", buckets.l2), ("L3", buckets.l3), ("L4/L5", buckets.l45)):
-        files = [path for platform in PLATFORMS for path in by_platform[platform]]
-        if files:
-            changed_parts.append(f"{level_name}=[{', '.join(files)}]")
-
-    run_targets = [
-        f"{platform}/{level}" for platform in PLATFORMS for level in ("l2", "l3") if decision.is_run(platform, level)
-    ]
-    skip_targets = [
-        f"{platform}/{level}"
-        for platform in PLATFORMS
-        for level in ("l2", "l3")
-        if not decision.is_run(platform, level)
-    ]
-
-    changed = "; ".join(changed_parts) if changed_parts else "none"
-    run = ", ".join(run_targets) if run_targets else "none"
-    skip = ", ".join(skip_targets) if skip_targets else "none"
-    if buckets.has_doc_changes or buckets.has_skip_test_changes:
-        return (
-            "CI config YAML + docs/skip-mark changed — follow CI YAML gating; "
-            f"changed: {changed}; run: {run}; skip: {skip}"
-        )
-    return f"only CI config YAML changed — changed: {changed}; run: {run}; skip: {skip}"
-
-
-def resolve_ci_decision(
-    changed_files: list[str] | None = None,
-    *,
-    diff_range: str | None = None,
-) -> CiDecision:
-    if changed_files is None:
-        return _finish(CiDecision(), "could not resolve changed files; run normal CI")
-
-    buckets = _classify_changed_files_into_buckets(changed_files, diff_range=diff_range)
-    if buckets.has_other_changes or not buckets.has_skip_ci_scope_changes:
-        return _finish(CiDecision(), "product or non-whitelisted changes present; run normal CI")
-
-    if CiDecision.skip_all_basis(buckets, diff_range=diff_range):
-        return _finish(
-            CiDecision(skip_all=True),
-            "CI skipped — docs or pytest skip-mark changes only",
-        )
-
-    # Docs / skip-mark mixed with whitelisted CI YAML → same yaml-gated path
-    # as CI-YAML-only (docs/skip-mark do not widen to normal CI).
-    if not buckets.has_l23_yaml_changes and not buckets.has_l45_yaml_changes:
-        if buckets.has_skip_test_changes and diff_range is None:
-            message = (
-                "only pytest skip-mark test changes detected, but no git diff range "
-                "is available to verify them (skip-all requires diff_range); run normal CI"
-            )
-        elif buckets.has_skip_test_changes:
-            message = "pytest skip-mark test changes did not qualify for skip-all and no ci yaml changed; run normal CI"
-        else:
-            message = "doc/skip-mark changes did not qualify for skip-all; run normal CI"
-        return _finish(CiDecision(), message)
-
-    decision = CiDecision.from_yaml_basis(
-        SkipL2L3Basis.from_buckets(buckets),
-        SkipL3Basis.from_buckets(buckets),
-        SkipL2Basis.from_buckets(buckets),
-    )
-    return _finish(decision, _format_yaml_gated_message(buckets, decision))
-
-
-def main() -> int:
-    parser = argparse.ArgumentParser(
-        description=__doc__,
-        formatter_class=argparse.RawDescriptionHelpFormatter,
-    )
-    subparsers = parser.add_subparsers(dest="command", required=True)
-
-    gate_parser = subparsers.add_parser(
-        "gate",
-        help="exit 0 when bootstrap should stop (skip-all or PLATFORM/LEVEL disabled); logs once",
-    )
-    gate_parser.add_argument("platform", help="cuda, amd, intel, or npu")
-    gate_parser.add_argument("level", choices=("l2", "l3"), help="test level")
-
-    subparsers.add_parser(
-        "check-skip-all",
-        help="exit 0 when docs/skip-mark-only diff should skip entire CI upload",
-    )
-
-    skip_parser = subparsers.add_parser(
-        "check-skip-l2-l3",
-        help="exit 0 when PLATFORM l2/l3 pipeline should be skipped (CI config YAML-only diffs)",
-    )
-    skip_parser.add_argument("platform", help="cuda, amd, intel, or npu")
-    skip_parser.add_argument("level", choices=("l2", "l3"), help="test level")
-
-    subparsers.add_parser(
-        "print-annotate",
-        help="print human-readable skip reason for Buildkite annotation",
-    )
-
-    args = parser.parse_args()
-    decision = resolve_ci_context_from_git().decision
-
-    if args.command == "gate":
-        if decision.skip_all:
-            print("skip-all")
-            return 0
-        if not decision.is_run(args.platform, args.level):
-            print("skip-l23")
-            return 0
-        return 1
-
-    if args.command == "check-skip-all":
-        return 0 if decision.skip_all else 1
-
-    if args.command == "check-skip-l2-l3":
-        return 0 if not decision.is_run(args.platform, args.level) else 1
-
-    if args.command == "print-annotate":
-        print(decision.message or "CI will run normally")
-        return 0
-
-    return 1
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())

.buildkite/common/scripts/upload_pipeline.py

@@ -1,449 +0,0 @@
-#!/usr/bin/env python3
-# SPDX-License-Identifier: Apache-2.0
-"""Render and optionally upload Buildkite pipeline YAML with diff-aware logic.
-
-Bootstrap mode (``bootstrap-upload-steps.yml``):
-  - Hook uploads the entry YAML (``pipeline.yml`` / ``pipeline-npu.yml``) with one step that runs
-    ``upload_pipeline.py --upload <platform>/bootstrap-upload-steps.yml``.
-  - Injects ``if`` by step ``key`` from skip-ci and uploads child steps (image build, L2–L5 upload).
-  - Detect docs-only, pytest skip-mark-only, or combined skip-ci from git diff.
-  - When only CI level YAML changes, enable **L2/L3** upload steps for affected levels only.
-
-Test pipeline mode (e.g. test-merge.yml):
-  - Drop steps whose ``source_file_dependencies`` do not match changed files.
-  - Expand uploader-only ``mirror_hardwares: l4_1`` into ``agents`` (+ optional ``image``
-    for NPU) + ``plugins`` (see ci_mirror_hardwares.yml).
-
-Usage:
-  python3 upload_pipeline.py [--upload] [--all | --e2e] <pipeline.yml>
-
-Requires PyYAML (``pip install pyyaml``); installs it automatically when missing.
-"""
-
-from __future__ import annotations
-
-import argparse
-import copy
-import subprocess
-import sys
-from functools import lru_cache
-from pathlib import Path
-from typing import Any
-
-try:
-    import yaml
-except ModuleNotFoundError:
-    subprocess.run(
-        [sys.executable, "-m", "pip", "install", "-q", "pyyaml"],
-        check=True,
-    )
-    import yaml
-
-from skip_ci import (
-    ROOT,
-    resolve_ci_context_from_git,
-)
-
-# --- Constants ---
-
-LOG = "upload_pipeline"
-BOOTSTRAP_STEPS_FILENAME = "bootstrap-upload-steps.yml"
-BOOTSTRAP_IMAGE_BUILD_KEYS = frozenset({"image-build", "image-build-a2", "image-build-a3"})
-BOOTSTRAP_UPLOAD_IF_KEYS = {
-    "upload-ready-pipeline": "ready",
-    "upload-merge-pipeline": "merge",
-    "upload-nightly-pipeline": "nightly",
-    "upload-weekly-pipeline": "weekly",
-}
-E2E_GROUP_MARKER = "E2E Test"
-CI_MIRROR_HARDWARES_PATH = ROOT / ".buildkite/common/ci_mirror_hardwares.yml"
-
-CUDA_NIGHTLY_ONLY = (
-    '(build.pull_request.labels includes "nightly-test") || (build.branch == "main" && build.env("NIGHTLY") == "1")'
-)
-NPU_NIGHTLY_ONLY = (
-    '(build.branch == "main" && build.env("NIGHTLY") == "1") || '
-    '(build.branch != "main" && ('
-    'build.pull_request.labels includes "nightly-test" || '
-    'build.pull_request.labels includes "omni-test" || '
-    'build.pull_request.labels includes "tts-test" || '
-    'build.pull_request.labels includes "diffusion-x2iat-test" || '
-    'build.pull_request.labels includes "diffusion-x2v-test"'
-    "))"
-)
-
-
-# --- Logging ---
-
-
-def _log(message: str) -> None:
-    print(f"{LOG}: {message}", file=sys.stderr)
-
-
-# --- Bootstrap pipeline (bootstrap-upload-steps.yml) ---
-
-
-def _get_bootstrap_platform(path: Path) -> str:
-    parts = path.as_posix().split("/")
-    return "npu" if "npu" in parts else "cuda"
-
-
-def _load_bootstrap_steps(path: Path) -> str:
-    if path.name != BOOTSTRAP_STEPS_FILENAME:
-        raise ValueError(f"expected {BOOTSTRAP_STEPS_FILENAME}, got {path.name}")
-    return path.read_text(encoding="utf-8")
-
-
-def _format_bootstrap_if(expr: str) -> str:
-    """Return a Buildkite ``if`` string. Buildkite rejects YAML bool ``if`` values."""
-    if expr in ("true", "false"):
-        return expr
-    return f"({expr})"
-
-
-def _compute_bootstrap_if_exprs(*, decision, platform: str) -> dict[str, str]:
-    disabled = "false"
-    nightly_main = 'build.branch == "main" && build.env("NIGHTLY") == "1"'
-    nightly_only = NPU_NIGHTLY_ONLY if platform == "npu" else CUDA_NIGHTLY_ONLY
-
-    if platform == "npu":
-        ready_pr = (
-            'build.branch != "main" && ('
-            'build.pull_request.labels includes "npu-test" || '
-            'build.pull_request.labels includes "ready"'
-            ")"
-        )
-        nightly_label_if = nightly_only
-        weekly_label_if = disabled
-        merge_base = disabled
-    else:
-        ready_pr = 'build.branch != "main" && build.pull_request.labels includes "ready"'
-        merge_main = 'build.branch == "main" && build.env("NIGHTLY") != "1" && build.env("WEEKLY") != "1"'
-        merge_pr = 'build.branch != "main" && build.pull_request.labels includes "merge-test"'
-        nightly_label_if = (
-            '(build.branch == "main" && build.env("NIGHTLY") == "1") || '
-            '(build.branch != "main" && ('
-            'build.pull_request.labels includes "nightly-test" || '
-            'build.pull_request.labels includes "omni-test" || '
-            'build.pull_request.labels includes "tts-test" || '
-            'build.pull_request.labels includes "diffusion-x2iat-test" || '
-            'build.pull_request.labels includes "diffusion-x2v-test"'
-            "))"
-        )
-        weekly_label_if = (
-            '(build.branch == "main" && build.env("WEEKLY") == "1") || '
-            '(build.branch != "main" && build.pull_request.labels includes "weekly-test")'
-        )
-        merge_base = f"({nightly_main}) || (({merge_main}) || ({merge_pr}))"
-
-    ready_base = f"({nightly_main}) || ({ready_pr})"
-    weekly_main = 'build.branch == "main" && build.env("WEEKLY") == "1"'
-
-    if decision.skip_all:
-        # Docs / skip-mark only: no PR-label escape hatch. Main scheduled
-        # NIGHTLY=1 still runs L4 + L2/L3 (--e2e); WEEKLY=1 still runs L5.
-        image_expr = f"({nightly_main}) || ({weekly_main})" if platform == "cuda" else nightly_main
-        ready_expr = nightly_main
-        merge_expr = nightly_main if platform == "cuda" else disabled
-        nightly_expr = nightly_main
-        weekly_expr = weekly_main if platform == "cuda" else disabled
-    elif decision.skip_l2_l3:
-        l2_enabled = decision.is_run("npu", "l2") if platform == "npu" else decision.is_run("cuda", "l2")
-        l3_enabled = platform == "cuda" and decision.is_run("cuda", "l3")
-
-        ready_expr = ready_base if l2_enabled else disabled
-        merge_expr = merge_base if l3_enabled else disabled
-        nightly_expr = nightly_label_if
-        weekly_expr = weekly_label_if if platform == "cuda" else disabled
-
-        image_parts = [f"({nightly_label_if})"]
-        if platform == "cuda":
-            image_parts.append(f"({weekly_label_if})")
-        if l2_enabled:
-            image_parts.insert(0, f"({ready_base})")
-        if l3_enabled:
-            image_parts.insert(1 if l2_enabled else 0, f"({merge_base})")
-        image_expr = " || ".join(image_parts)
-    else:
-        image_expr = "true"
-        ready_expr = ready_base
-        merge_expr = merge_base if platform == "cuda" else disabled
-        nightly_expr = nightly_label_if
-        weekly_expr = weekly_label_if if platform == "cuda" else disabled
-
-    return {
-        "image": _format_bootstrap_if(image_expr),
-        "ready": _format_bootstrap_if(ready_expr),
-        "merge": _format_bootstrap_if(merge_expr),
-        "nightly": _format_bootstrap_if(nightly_expr),
-        "weekly": _format_bootstrap_if(weekly_expr),
-    }
-
-
-def _apply_bootstrap_if(steps: list[Any], if_exprs: dict[str, str]) -> list[Any]:
-    """Inject ``if`` by step key; drop steps that are unconditionally disabled."""
-    kept: list[Any] = []
-    for step in steps:
-        if not isinstance(step, dict):
-            kept.append(step)
-            continue
-        nested = step.get("steps")
-        if isinstance(nested, list):
-            nested_kept = _apply_bootstrap_if(nested, if_exprs)
-            if nested_kept:
-                kept.append({**step, "steps": nested_kept})
-            continue
-        key = step.get("key")
-        if key in BOOTSTRAP_IMAGE_BUILD_KEYS:
-            step["if"] = if_exprs["image"]
-        elif key in BOOTSTRAP_UPLOAD_IF_KEYS:
-            step["if"] = if_exprs[BOOTSTRAP_UPLOAD_IF_KEYS[key]]
-        if_expr = step.get("if")
-        if if_expr == "false":
-            _log(f"omit disabled bootstrap step {key!r}")
-            continue
-        if if_expr == "true":
-            # Unconditional step: omit ``if`` (Buildkite requires string ``if``, not YAML bool).
-            step.pop("if", None)
-        kept.append(step)
-    return kept
-
-
-def _render_bootstrap_pipeline(
-    steps_yaml: str,
-    *,
-    decision,
-    path: Path,
-) -> str:
-    """Load bootstrap steps YAML and inject ``if`` expressions from skip-ci decision."""
-    doc = yaml.safe_load(steps_yaml)
-    if not isinstance(doc, dict):
-        raise ValueError(f"invalid bootstrap steps YAML for {path}")
-    steps = doc.get("steps")
-    if not isinstance(steps, list):
-        raise ValueError(f"bootstrap steps YAML must contain steps: list in {path}")
-
-    platform = _get_bootstrap_platform(path)
-    if_exprs = _compute_bootstrap_if_exprs(decision=decision, platform=platform)
-    doc["steps"] = _apply_bootstrap_if(steps, if_exprs)
-    return yaml.safe_dump(doc, sort_keys=False)
-
-
-# --- Test pipeline (test-ready.yml, test-merge.yml) ---
-
-
-@lru_cache(maxsize=1)
-def _load_mirror_hardwares() -> dict[str, dict[str, Any]]:
-    if not CI_MIRROR_HARDWARES_PATH.is_file():
-        raise FileNotFoundError(f"missing CI mirror_hardwares registry: {CI_MIRROR_HARDWARES_PATH}")
-    doc = yaml.safe_load(CI_MIRROR_HARDWARES_PATH.read_text(encoding="utf-8"))
-    if not isinstance(doc, dict):
-        raise ValueError(f"invalid CI mirror_hardwares registry: {CI_MIRROR_HARDWARES_PATH}")
-    presets = doc.get("mirror_hardwares")
-    if not isinstance(presets, dict):
-        raise ValueError(f"mirror_hardwares must be a mapping in {CI_MIRROR_HARDWARES_PATH}")
-    return presets
-
-
-def _expand_mirror_hardwares(step: dict[str, Any]) -> dict[str, Any]:
-    """Replace uploader-only ``mirror_hardwares`` with preset fields from ci_mirror_hardwares.yml."""
-    hardware = step.get("mirror_hardwares")
-    if hardware is None:
-        return step
-
-    if not isinstance(hardware, str) or not hardware.strip():
-        raise ValueError(
-            f"mirror_hardwares must be a non-empty string in step {_get_step_label(step)!r}",
-        )
-
-    preset = _load_mirror_hardwares().get(hardware)
-    if preset is None:
-        known = ", ".join(sorted(_load_mirror_hardwares()))
-        raise ValueError(
-            f"unknown mirror_hardwares {hardware!r} in step {_get_step_label(step)!r}; known: {known}",
-        )
-
-    if step.get("agents") is not None or step.get("plugins") is not None or step.get("image") is not None:
-        raise ValueError(
-            f"step {_get_step_label(step)!r} sets mirror_hardwares together with agents/plugins/image; "
-            "use mirror_hardwares only",
-        )
-
-    expanded = copy.deepcopy(preset)
-    return {key: value for key, value in step.items() if key != "mirror_hardwares"} | expanded
-
-
-def _match_source_file(changed_files: list[str], prefixes: list[str]) -> bool:
-    for path in changed_files:
-        for prefix in prefixes:
-            normalized = prefix.rstrip("/")
-            if path == normalized or path.startswith(f"{normalized}/"):
-                return True
-    return False
-
-
-def _get_step_label(step: dict[str, Any]) -> str:
-    return str(step.get("group") or step.get("label") or "<step>")
-
-
-def _process_test_steps(
-    steps: list[Any],
-    changed_files: list[str] | None,
-) -> list[Any]:
-    """Drop steps by ``source_file_dependencies`` when *changed_files* is set; always strip that field."""
-    processed: list[Any] = []
-    for step in steps:
-        if not isinstance(step, dict):
-            processed.append(step)
-            continue
-
-        deps = step.get("source_file_dependencies")
-        if deps is not None and not isinstance(deps, list):
-            raise ValueError(
-                f"source_file_dependencies must be a list in step {_get_step_label(step)!r}",
-            )
-        if changed_files is not None and deps is not None and not _match_source_file(changed_files, deps):
-            _log(f"skip {_get_step_label(step)!r} (no changes under {deps})")
-            continue
-
-        nested = step.get("steps")
-        if nested is not None:
-            kept_nested = _process_test_steps(nested, changed_files)
-            if changed_files is not None and not kept_nested:
-                _log(f"omit empty group {_get_step_label(step)!r}")
-                continue
-            new_step = {key: value for key, value in step.items() if key != "source_file_dependencies"}
-            new_step["steps"] = kept_nested
-            processed.append(new_step)
-            continue
-
-        if deps is not None:
-            processed.append(
-                _expand_mirror_hardwares(
-                    {key: value for key, value in step.items() if key != "source_file_dependencies"},
-                ),
-            )
-        else:
-            processed.append(_expand_mirror_hardwares(step))
-
-    return processed
-
-
-def _select_e2e_group_steps(steps: list[Any]) -> list[Any]:
-    """Keep only top-level groups whose name contains ``E2E_GROUP_MARKER``."""
-    selected = [
-        step
-        for step in steps
-        if isinstance(step, dict) and isinstance(step.get("group"), str) and E2E_GROUP_MARKER in step["group"]
-    ]
-    if not selected:
-        _log(f"no group matching {E2E_GROUP_MARKER!r} found")
-    else:
-        _log(f"keep {len(selected)} group(s) matching {E2E_GROUP_MARKER!r}")
-    return selected
-
-
-def _render_test_pipeline(
-    doc: dict[str, Any],
-    changed_files: list[str] | None,
-    *,
-    e2e_only: bool = False,
-) -> dict[str, Any]:
-    """Filter steps by PR diff and strip uploader-only ``source_file_dependencies`` metadata."""
-    steps = doc.get("steps")
-    if not isinstance(steps, list):
-        return doc
-    if e2e_only:
-        steps = _select_e2e_group_steps(steps)
-    steps = _process_test_steps(steps, changed_files)
-    return {**doc, "steps": steps}
-
-
-# --- Entry (read file → bootstrap or test render → YAML string) ---
-
-
-def _render_pipeline(
-    path: Path,
-    *,
-    force_all: bool = False,
-    e2e_only: bool = False,
-) -> str:
-    if path.name == BOOTSTRAP_STEPS_FILENAME:
-        ctx = resolve_ci_context_from_git()
-        continuation = _load_bootstrap_steps(path)
-        return _render_bootstrap_pipeline(
-            continuation,
-            decision=ctx.decision,
-            path=path,
-        )
-
-    text = path.read_text(encoding="utf-8")
-    ctx = resolve_ci_context_from_git()
-    if force_all or e2e_only:
-        changed_files = None
-    else:
-        changed_files = ctx.changed_files
-
-    doc = yaml.safe_load(text)
-    if not isinstance(doc, dict):
-        raise ValueError(f"invalid pipeline YAML: {path}")
-
-    doc = _render_test_pipeline(doc, changed_files, e2e_only=e2e_only)
-    return yaml.safe_dump(doc, sort_keys=False)
-
-
-def _upload_to_buildkite(content: str) -> None:
-    subprocess.run(
-        ["buildkite-agent", "pipeline", "upload"],
-        input=content,
-        text=True,
-        check=True,
-    )
-
-
-# --- CLI ---
-
-
-def main() -> int:
-    parser = argparse.ArgumentParser(description=__doc__)
-    parser.add_argument(
-        "pipeline",
-        nargs="?",
-        default=".buildkite/cuda/bootstrap-upload-steps.yml",
-        help="Pipeline YAML path (default: .buildkite/cuda/bootstrap-upload-steps.yml)",
-    )
-    parser.add_argument(
-        "--upload",
-        action="store_true",
-        help="Pipe rendered YAML to buildkite-agent pipeline upload",
-    )
-    mode = parser.add_mutually_exclusive_group()
-    mode.add_argument(
-        "--all",
-        action="store_true",
-        help="Keep all steps (disable diff-aware skipping)",
-    )
-    mode.add_argument(
-        "--e2e",
-        action="store_true",
-        help="Keep only the E2E Test group",
-    )
-    args = parser.parse_args()
-
-    path = Path(args.pipeline)
-    if not path.is_absolute():
-        path = ROOT / path
-    if not path.is_file():
-        _log(f"missing pipeline file: {path}")
-        return 1
-
-    rendered = _render_pipeline(path, force_all=args.all, e2e_only=args.e2e)
-    if args.upload:
-        _upload_to_buildkite(rendered)
-    else:
-        sys.stdout.write(rendered)
-    return 0
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())

.buildkite/cuda/bootstrap-upload-steps.yml

@@ -1,55 +0,0 @@
-# Uploaded by upload_pipeline.py --upload (not the Buildkite hook).
-# ``if`` is injected by step ``key`` from skip-ci at upload time.
-steps:
-  - label: ":docker: Build image"
-    key: image-build
-    commands:
-      - "bash .buildkite/cuda/scripts/build-ci-image.sh"
-    agents:
-      queue: "cpu_queue_premerge"
-
-  # L2 Test — PR ready label (diff-aware) or main+NIGHTLY=1 (--e2e, full E2E group)
-  - label: "Upload Ready Pipeline"
-    depends_on: image-build
-    key: upload-ready-pipeline
-    commands:
-      - |
-        EXTRA=""
-        if [ "${NIGHTLY:-}" = "1" ] && [ "${BUILDKITE_BRANCH}" = "main" ]; then
-          EXTRA="--e2e"
-        fi
-        python3 .buildkite/common/scripts/upload_pipeline.py --upload $$EXTRA .buildkite/cuda/test-ready.yml
-    agents:
-      queue: "cpu_queue_premerge"
-
-  # L3 Test — PR merge-test label (diff-aware) or main+NIGHTLY=1 (--e2e, full E2E group)
-  - label: "Upload Merge Pipeline"
-    depends_on: image-build
-    key: upload-merge-pipeline
-    commands:
-      - |
-        EXTRA=""
-        if [ "${NIGHTLY:-}" = "1" ] && [ "${BUILDKITE_BRANCH}" = "main" ]; then
-          EXTRA="--e2e"
-        fi
-        python3 .buildkite/common/scripts/upload_pipeline.py --upload $$EXTRA .buildkite/cuda/test-merge.yml
-    agents:
-      queue: "cpu_queue_premerge"
-
-  # L4 Test — main+NIGHTLY=1 (scheduled), or PR with specific label (e.g. add label then Rebuild)
-  - label: "Upload Nightly Pipeline"
-    depends_on: image-build
-    key: upload-nightly-pipeline
-    commands:
-      - python3 .buildkite/common/scripts/upload_pipeline.py --upload .buildkite/cuda/test-nightly.yml
-    agents:
-      queue: "cpu_queue_premerge"
-
-  # L5 Test — main+WEEKLY=1 (scheduled), or PR with weekly-test label
-  - label: "Upload Weekly Pipeline"
-    depends_on: image-build
-    key: upload-weekly-pipeline
-    commands:
-      - python3 .buildkite/common/scripts/upload_pipeline.py --upload .buildkite/cuda/test-weekly.yml
-    agents:
-      queue: "cpu_queue_premerge"

.buildkite/cuda/pipeline.yml

@@ -1,9 +0,0 @@
-# Hook uploads this entry YAML; upload_pipeline.py --upload reads bootstrap-upload-steps.yml
-# and injects ``if`` by step ``key`` from skip-ci.
-steps:
-  - label: ":github: Resolve skip-ci & upload pipeline"
-    key: upload-ci-pipeline
-    commands:
-      - "python3 .buildkite/common/scripts/upload_pipeline.py --upload .buildkite/cuda/bootstrap-upload-steps.yml"
-    agents:
-      queue: "cpu_queue_premerge"

.buildkite/cuda/scripts/build-ci-image.sh

@@ -1,52 +0,0 @@
-#!/bin/bash
-set -euo pipefail
-
-# Build and push the CI Docker image with registry-based layer caching.
-#
-# Usage: build-ci-image.sh
-#
-# Expects the following environment variables (set by Buildkite):
-#   BUILDKITE_COMMIT          - commit SHA used as the image tag
-ECR_NAMESPACE="public.ecr.aws/q9t5s3a7"
-REGISTRY="${ECR_NAMESPACE}/vllm-ci-test-repo"
-REGION="us-east-1"
-DOCKERFILE="docker/Dockerfile.ci"
-BUILDER_NAME="vllm-omni-builder"
-
-# Ensure that the env vars are actually set, otherwise exit early
-if [ -z "${BUILDKITE_COMMIT:-}" ]; then
-    echo "ERROR: BUILDKITE_COMMIT is not set"
-    exit 1
-fi
-echo "BUILDKITE_COMMIT: ${BUILDKITE_COMMIT}"
-
-# Authenticate to ECR Public
-aws ecr-public get-login-password --region "$REGION" \
-    | docker login --username AWS --password-stdin "$ECR_NAMESPACE"
-
-# Compute the cache tag for the dependencies image as the hash of the contents.
-# This is because dependencies don't change often, so most PRs will have the same hash.
-#
-# The Dockerfile is also included to avoid churn in the cached image based on the vLLM version.
-DEP_FILES="pyproject.toml setup.py requirements/*.txt docker/Dockerfile.ci"
-CACHE_KEY=$(cat ${DEP_FILES} | sha256sum | cut -c1-16)
-CACHE_TAG="deps-cache-${CACHE_KEY}"
-echo "Cache key: ${CACHE_TAG}"
-
-# Set up buildx with docker-container driver; we need to do this
-# since cache export is not supported for the default docker driver
-# that is running in the CI.
-docker buildx inspect $BUILDER_NAME >/dev/null 2>&1 \
-|| docker buildx create --name $BUILDER_NAME --driver docker-container
-docker buildx use $BUILDER_NAME
-
-echo "Building image tag: ${REGISTRY}:${BUILDKITE_COMMIT}"
-
-# Build + push the image to the registry, using the hash from the dependencies.
-# Note that the caching here is layerwise, so we're just making sure we always
-# can cache hit on everything but the last copy at the moment.
-docker buildx build --push --progress=plain \
-    --cache-from "type=registry,ref=${REGISTRY}:${CACHE_TAG}" \
-    --cache-to "type=registry,ref=${REGISTRY}:${CACHE_TAG},mode=max,compression=zstd" \
-    --file "$DOCKERFILE" \
-    -t "${REGISTRY}:${BUILDKITE_COMMIT}" .

.buildkite/cuda/test-merge.yml

@@ -1,259 +0,0 @@
-env:
-  VLLM_WORKER_MULTIPROC_METHOD: spawn
-  VLLM_USE_FLASHINFER_MOE_FP16: 0
-  HF_HUB_DOWNLOAD_TIMEOUT: 300
-  HF_HUB_ETAG_TIMEOUT: 60
-
-steps:
-  - group: ":card_index_dividers: Simple Test"
-    depends_on: upload-merge-pipeline
-    steps:
-      - label: "Simple · Model Executor Test"
-        commands:
-          - "timeout 20m pytest -sv tests/model_executor -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-        mirror_hardwares: l4_1
-
-      - label: "Simple · Diffusion Test"
-        commands:
-          - "timeout 20m pytest -sv tests/diffusion -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-        mirror_hardwares: l4_1
-
-      - label: "Simple · Engine&Entrypoints Test"
-        commands:
-          - "timeout 20m pytest -sv tests/entrypoints tests/engine -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-        mirror_hardwares: l4_1
-
-      - label: "Simple · Other Test"
-        commands:
-          - "timeout 20m pytest -sv tests/ -m 'core_model and cpu' --ignore=tests/diffusion --ignore=tests/model_executor --ignore=tests/entrypoints --ignore=tests/engine --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-        mirror_hardwares: l4_1
-
-  - group: ":card_index_dividers: Diffusion Test"
-    depends_on: upload-merge-pipeline
-    steps:
-      - label: "Diffusion · LoRA Test"
-        timeout_in_minutes: 30
-        commands:
-          - pytest -sv tests/diffusion/lora/ -m 'advanced_model and cuda' --run-level "advanced_model"
-        mirror_hardwares: l4_1
-
-  - label: "Distributed Test"
-    depends_on: upload-merge-pipeline
-    timeout_in_minutes: 60
-    commands:
-      - |
-        timeout 55m bash -c "
-          set -e
-          export VLLM_IMAGE_FETCH_TIMEOUT=60
-          pytest -s -v tests/distributed -m 'advanced_model and cuda and H100' --run-level 'advanced_model'
-        "
-    mirror_hardwares: h100_1
-
-  - group: ":card_index_dividers: Entrypoints Test"
-    depends_on: upload-merge-pipeline
-    steps:
-      - label: "Entrypoint Test with H100"
-        # TEMPORARY 60 -> 90: engine startup/sleep-wake is ~13-15 min slower per
-        # run on the vLLM 0.26 image (model loading ~2x, init ~2x; see builds
-        # 2915 vs 2916), which pushed this suite past 60 min. Revert once the
-        # 0.26 startup regression is understood and fixed.
-        timeout_in_minutes: 90
-        commands:
-          - pytest -s -v tests/entrypoints/ -m "advanced_model and cuda and H100" --run-level "advanced_model"
-        mirror_hardwares: h100_2
-
-      - label: "Entrypoint Test with L4"
-        timeout_in_minutes: 60
-        commands:
-          - pytest -s -v tests/entrypoints/ -m "advanced_model and cuda and L4" --run-level "advanced_model"
-        mirror_hardwares: l4_1
-
-  - group: ":card_index_dividers: E2E Test"
-    depends_on: upload-merge-pipeline
-    steps:
-      - label: "TTS · Qwen3-TTS CustomVoice Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen3_tts_customvoice.py
-          - tests/e2e/offline_inference/test_qwen3_tts_customvoice.py
-          - vllm_omni/model_executor/models/qwen3_tts/
-          - vllm_omni/model_executor/stage_input_processors/qwen3_tts.py
-          - vllm_omni/deploy/qwen3_tts.yaml
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_LOGGING_LEVEL=DEBUG
-              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
-              pytest -s -v tests/e2e/online_serving/test_qwen3_tts_customvoice.py tests/e2e/offline_inference/test_qwen3_tts_customvoice.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: "TTS · Qwen3-TTS Base Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen3_tts_base.py
-          - tests/e2e/offline_inference/test_qwen3_tts_base.py
-          - vllm_omni/model_executor/models/qwen3_tts/
-          - vllm_omni/model_executor/stage_input_processors/qwen3_tts.py
-          - vllm_omni/deploy/qwen3_tts.yaml
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_LOGGING_LEVEL=DEBUG
-              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
-              pytest -s -v tests/e2e/online_serving/test_qwen3_tts_base.py tests/e2e/offline_inference/test_qwen3_tts_base.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: "TTS · Higgs-Audio-V3 Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_higgs_audio_v3.py
-          - vllm_omni/model_executor/models/higgs_audio_v3/
-          - vllm_omni/model_executor/stage_input_processors/higgs_audio_v3.py
-          - vllm_omni/deploy/higgs_multimodal_qwen3.yaml
-        commands:
-          - |
-            timeout 20m bash -c "
-              export VLLM_WORKER_MULTIPROC_METHOD=spawn
-              export VLLM_USE_DEEP_GEMM=0
-              export VLLM_MOE_USE_DEEP_GEMM=0
-              pytest -s -v tests/e2e/online_serving/test_higgs_audio_v3.py -m 'advanced_model' --run-level 'advanced_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: "Omni · Qwen3-Omni Test"
-        source_file_dependencies:
-          - tests/e2e/offline_inference/test_qwen3_omni.py
-          - tests/e2e/online_serving/test_qwen3_omni.py
-          - vllm_omni/model_executor/models/qwen3_omni/
-          - vllm_omni/model_executor/stage_input_processors/qwen3_omni.py
-          - vllm_omni/deploy/qwen3_omni_moe.yaml
-        timeout_in_minutes: 50
-        commands:
-          - pytest -s -v tests/e2e/offline_inference/test_qwen3_omni.py tests/e2e/online_serving/test_qwen3_omni.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-        mirror_hardwares: h100_2
-
-      - label: "Omni · MiniCPM-o 4.5 Test"
-        source_file_dependencies:
-          - tests/e2e/offline_inference/test_minicpmo_4_5.py
-          - tests/e2e/online_serving/test_minicpmo_4_5.py
-          - vllm_omni/deploy/minicpmo_4_5.yaml
-        timeout_in_minutes: 50
-        commands:
-          - pytest -s -v tests/e2e/offline_inference/test_minicpmo_4_5.py tests/e2e/online_serving/test_minicpmo_4_5.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-        mirror_hardwares: h100_2
-
-      - label: "Omni · MiniCPM-o 4.5 Duplex Test"
-        source_file_dependencies:
-          - tests/assets/minicpmo_4_5/response_required_16k.wav
-          - tests/assets/minicpmo_4_5/soft_interrupt_16k.wav
-          - tests/e2e/online_serving/test_minicpmo_4_5_duplex.py
-          - tests/e2e/online_serving/test_minicpmo_4_5_duplex_expansion.py
-          - tests/e2e/online_serving/test_minicpmo_realtime_duplex_drivers.py
-          - tests/e2e/online_serving/minicpmo_realtime_duplex_scenarios.py
-          - tests/e2e/online_serving/run_minicpmo_realtime_duplex_multi_session.py
-          - tests/e2e/online_serving/run_minicpmo_realtime_duplex_soft_interrupt.py
-          - tests/helpers/minicpmo_4_5_duplex.py
-          - vllm_omni/deploy/minicpmo_4_5_duplex.yaml
-          - vllm_omni/experimental/fullduplex/
-          - vllm_omni/model_executor/models/minicpmo_4_5/
-          - vllm_omni/model_executor/stage_input_processors/minicpmo_4_5_omni.py
-          - vllm_omni/core/sched/
-          - vllm_omni/outputs/multimodal_accumulation.py
-          - vllm_omni/engine/stage_init_utils.py
-          - vllm_omni/engine/async_omni_engine.py
-          - vllm_omni/entrypoints/cli/serve.py
-        timeout_in_minutes: 50
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_minicpmo_4_5_duplex.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-        mirror_hardwares: h100_2
-
-      - label: "TTS · VoxCPM2 Test"
-        source_file_dependencies:
-          - tests/e2e/offline_inference/test_voxcpm2_tts.py
-          - tests/e2e/online_serving/test_voxcpm2_tts.py
-          - vllm_omni/model_executor/models/voxcpm2/
-          - vllm_omni/deploy/voxcpm2.yaml
-        timeout_in_minutes: 30
-        commands:
-          - |
-            timeout 30m bash -c "
-              export VLLM_LOGGING_LEVEL=DEBUG
-              export VLLM_WORKER_MULTIPROC_METHOD=spawn
-              pytest -s -v tests/e2e/offline_inference/test_voxcpm2_tts.py tests/e2e/online_serving/test_voxcpm2_tts.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: "Diffusion · Qwen Image Edit Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen_image_edit.py
-          - vllm_omni/diffusion/models/qwen_image/
-        timeout_in_minutes: 20
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_qwen_image_edit.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-        mirror_hardwares: h100_1
-      - label: "Diffusion · Bagel Test"
-        source_file_dependencies:
-          - tests/e2e/offline_inference/test_bagel.py
-          - tests/e2e/online_serving/test_bagel.py
-          - vllm_omni/model_executor/models/bagel/
-          - vllm_omni/diffusion/models/bagel/
-          - vllm_omni/model_executor/stage_input_processors/bagel.py
-          - vllm_omni/deploy/bagel.yaml
-        timeout_in_minutes: 20
-        commands:
-          - pytest -s -v tests/e2e/offline_inference/test_bagel.py tests/e2e/online_serving/test_bagel.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Qwen Image Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen_image.py
-          - vllm_omni/diffusion/models/qwen_image/
-        commands:
-          - |
-            timeout 40m bash -c '
-              export VLLM_IMAGE_FETCH_TIMEOUT=60
-              pytest -s -v tests/e2e/online_serving/test_qwen_image.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-            '
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Qwen Image Layered Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen_image_layered.py
-          - vllm_omni/diffusion/models/qwen_image/
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_IMAGE_FETCH_TIMEOUT=60
-              pytest -s -v tests/e2e/online_serving/test_qwen_image_layered.py -m 'advanced_model and cuda' --run-level 'advanced_model'
-            "
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Wan22 Test"
-        source_file_dependencies:
-          - tests/e2e/offline_inference/test_wan22_t2v.py
-          - tests/e2e/online_serving/test_wan22_t2v.py
-          - vllm_omni/diffusion/models/wan2_2/
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_IMAGE_FETCH_TIMEOUT=60
-              pytest -s -v tests/e2e/offline_inference/test_wan22_t2v.py tests/e2e/online_serving/test_wan22_t2v.py -m 'advanced_model and cuda and diffusion' --run-level 'advanced_model'
-            "
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Z Image Test"
-        source_file_dependencies:
-          - tests/e2e/offline_inference/test_z_image.py
-          - tests/e2e/online_serving/test_z_image.py
-          - vllm_omni/diffusion/models/z_image/
-        timeout_in_minutes: 30
-        commands:
-          - pytest -s -v tests/e2e/offline_inference/test_z_image.py tests/e2e/online_serving/test_z_image.py -m 'advanced_model and cuda and diffusion' --run-level 'advanced_model'
-        mirror_hardwares: l4_1
-      - label: "Diffusion · FLUX.2 Klein Test"
-        source_file_dependencies:
-          - tests/e2e/offline_inference/test_flux2_klein.py
-          - vllm_omni/diffusion/models/flux2_klein/
-        timeout_in_minutes: 30
-        commands:
-          - pytest -s -v tests/e2e/offline_inference/test_flux2_klein.py -m 'advanced_model and cuda and diffusion' --run-level 'advanced_model'
-        mirror_hardwares: l4_1

.buildkite/cuda/test-nightly.yml

@@ -1,557 +0,0 @@
-env:
-  VLLM_WORKER_MULTIPROC_METHOD: spawn
-  VLLM_USE_FLASHINFER_MOE_FP16: 0
-  HF_HUB_DOWNLOAD_TIMEOUT: 300
-  HF_HUB_ETAG_TIMEOUT: 60
-
-steps:
-  # Group: collapses under one heading in the Buildkite UI; child steps still run in parallel.
-  - group: ":card_index_dividers: Omni Model Test"
-    key: nightly-omni-test-group
-    depends_on: upload-nightly-pipeline
-    if: build.env("NIGHTLY") == "1" || build.pull_request.labels includes "nightly-test" || build.pull_request.labels includes "omni-test"
-    steps:
-
-      - label: ":full_moon: Omni · Function Test with H100"
-        timeout_in_minutes: 120
-        commands:
-          - pytest -sv tests/e2e --run-level "full_model" -m "full_model and H100 and omni" --ignore=tests/e2e/accuracy --ignore=tests/e2e/online_serving/test_qwen3_omni_multi_replicas.py --ignore=tests/e2e/online_serving/test_minicpmo_4_5_duplex_expansion.py
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Omni · MiniCPM-o 4.5 Duplex Test"
-        timeout_in_minutes: 50
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_minicpmo_4_5_duplex_expansion.py -m "full_model and cuda and H100 and omni" --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Omni · Doc Test with L4"
-        timeout_in_minutes: 90
-        commands:
-          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-          - pytest -s -v tests/examples/ -m "full_model and omni and L4" --run-level "full_model"
-        mirror_hardwares: l4_4
-
-      - label: ":full_moon: Omni · Doc Test with H100"
-        timeout_in_minutes: 90
-        commands:
-          - pytest -s -v tests/examples/ -m "full_model and omni and H100" --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Omni · Accuracy Test"
-        timeout_in_minutes: 180
-        commands:
-          - export SEED_TTS_WER_EVAL=1
-          - export SEED_TTS_EVAL_DEVICE=cuda:1
-          - |
-            set +e
-            pytest -s -v tests/e2e/accuracy/qwen3_omni/test_qwen3_omni.py -m "full_model" --run-level full_model
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/e2e/accuracy/qwen3_omni/results/qwen_omni_acc/*.json"
-            exit $$EXIT
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Omni · Perf Test · No Async Chunk"
-        key: nightly-omni-performance-no-async-chunk
-        timeout_in_minutes: 300
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen3_omni_no_async_chunk.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Omni · Perf Test · Async Chunk"
-        key: nightly-omni-performance-async-chunk
-        timeout_in_minutes: 180
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen3_omni_async_chunk.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Omni · Perf Test · vLLM Text"
-        key: nightly-omni-performance-vllm-text
-        timeout_in_minutes: 180
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen3_omni_vllm_text.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Omni · Perf Test · Multi-Replica"
-        key: nightly-omni-performance-multi-replicas
-        timeout_in_minutes: 180
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen3_omni_multi_replicas.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-        mirror_hardwares: h100_3
-
-      - label: ":full_moon: Omni · Multi-Replica Startup Test with 4x H100"
-        timeout_in_minutes: 45
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_qwen3_omni_multi_replicas.py -m "full_model" --run-level "core_model"
-        mirror_hardwares: h100_4
-
-  - group: ":card_index_dividers: TTS Model Test"
-    key: nightly-tts-test-group
-    depends_on: upload-nightly-pipeline
-    if: build.env("NIGHTLY") == "1" || build.pull_request.labels includes "nightly-test" || build.pull_request.labels includes "tts-test"
-    steps:
-
-      - label: ":full_moon: TTS · Function Test with L4"
-        timeout_in_minutes: 120
-        commands:
-          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-          - export VLLM_USE_DEEP_GEMM="0"
-          - export VLLM_MOE_USE_DEEP_GEMM="0"
-          - pytest -s -v tests/e2e/ -m "full_model and L4 and tts" --run-level "full_model" --ignore=tests/e2e/accuracy --ignore=tests/e2e/online_serving/test_voxcpm2_tts_expansion.py
-        mirror_hardwares: l4_1
-
-      - label: ":full_moon: TTS · Accuracy Test · VoxCPM2"
-        timeout_in_minutes: 30
-        commands:
-          - |
-            timeout 30m bash -c "
-              export VLLM_LOGGING_LEVEL=DEBUG
-              export VLLM_WORKER_MULTIPROC_METHOD=spawn
-              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
-              pytest -s -v tests/e2e/online_serving/test_voxcpm2_tts_expansion.py -m 'full_model and cuda and tts' --run-level 'full_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: ":full_moon: TTS · Perf Test"
-        key: nightly-tts-performance
-        timeout_in_minutes: 180
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_tts.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: TTS · VoxCPM2 · Perf Test"
-        key: nightly-tts-performance-voxcpm2
-        timeout_in_minutes: 180
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_voxcpm2.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-        mirror_hardwares: h100_1
-
-      - label: ":full_moon: TTS · Higgs-Audio-V3 · Perf Test"
-        key: nightly-tts-performance-higgs-audio-v3
-        timeout_in_minutes: 180
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - export VLLM_USE_DEEP_GEMM="0"
-          - export VLLM_MOE_USE_DEEP_GEMM="0"
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_higgs_audio_v3.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-        mirror_hardwares: h100_1
-
-  - group: ":card_index_dividers: Tiny Model Tests [Multi-GPU]"
-    key: nightly-tiny-model-test-group
-    depends_on: upload-nightly-pipeline
-    if: >-
-      build.env("NIGHTLY") == "1" ||
-      build.pull_request.labels includes "nightly-test"
-    steps:
-
-      - label: ":full_moon: Tiny Model Tests · Diffusion (Multi-GPU accelerations)"
-        timeout_in_minutes: 30
-        # NOTE: Single-GPU (core_model) tiny model tests run on the ready label,
-        # so nightly only adds the multi-GPU parallelism configurations.
-        #
-        # This is a bit confusing, but `full_model` is used to mark the multigpu tests, and
-        # --run-level `core_model` is used to indicate that we should use tiny model weights
-        # for the CI, since these tests should pass irrespective of whether or not we use
-        # the full weights or not.
-        commands:
-          - timeout 25m pytest -sv tests/model_tests/diffusion -m 'full_model and cuda' --run-level "core_model"
-        mirror_hardwares: l4_4
-
-  - group: ":card_index_dividers: Diffusion X2I(&A&T) Model Test"
-    key: nightly-diffusion-x2iat-group
-    depends_on: upload-nightly-pipeline
-    if: >-
-      build.env("NIGHTLY") == "1" ||
-      build.pull_request.labels includes "nightly-test" ||
-      build.pull_request.labels includes "diffusion-x2iat-test"
-    steps:
-      # Use explicit file shards so single-card cases do not reserve 2+ GPUs,
-      # and Bagel feature coverage is not duplicated via a broad tests/e2e sweep.
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with H100 · Single-GPU"
-        timeout_in_minutes: 120
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/online_serving/test_qwen_image_expansion.py
-            tests/e2e/online_serving/test_qwen_image_edit_expansion.py
-            tests/e2e/online_serving/test_qwen_image_layered_expansion.py
-            tests/e2e/online_serving/test_boogu_image.py
-            tests/e2e/online_serving/test_boogu_image_edit.py
-            tests/e2e/online_serving/test_boogu_image_expansion.py
-            tests/e2e/online_serving/test_boogu_image_edit_expansion.py
-            tests/e2e/online_serving/test_bagel_expansion.py
-            tests/e2e/online_serving/test_krea2_expansion.py
-            tests/e2e/offline_inference/test_krea2_expansion.py
-            -m "full_model and diffusion and H100 and not distributed_cuda"
-            --run-level "full_model"
-        mirror_hardwares: h100_1
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with H100 · Multi-GPU Qwen-Image"
-        timeout_in_minutes: 120
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/online_serving/test_qwen_image_expansion.py
-            tests/e2e/online_serving/test_qwen_image_edit_expansion.py
-            tests/e2e/offline_inference/test_krea2_expansion.py
-            -m "full_model and diffusion and H100 and distributed_cuda"
-            --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with H100 · Multi-GPU Layered/LongCat/FLUX"
-        timeout_in_minutes: 90
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/online_serving/test_qwen_image_layered_expansion.py
-            -m "full_model and diffusion and H100 and distributed_cuda"
-            --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with H100 · Multi-GPU Bagel"
-        timeout_in_minutes: 120
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/online_serving/test_bagel_expansion.py
-            -m "full_model and diffusion and H100 and distributed_cuda"
-            --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · GLM-Image Function Test with H100"
-        timeout_in_minutes: 120
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/online_serving/test_glm_image_expansion.py
-            tests/e2e/offline_inference/test_glm_image_autoround_w4a16_expansion.py
-            -m "full_model and diffusion and H100"
-            --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with L4"
-        timeout_in_minutes: 120
-        commands:
-          - pytest -sv tests/e2e/ -k "not test_wan and not test_bagel_expansion and not hunyuan" -m "full_model and diffusion and L4" --run-level "full_model" --ignore=tests/e2e/accuracy
-        mirror_hardwares: l4_4
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Doc Test"
-        timeout_in_minutes: 60
-        commands:
-          - pytest -s -v tests/examples/*/test_text_to_image.py -m "full_model and example and H100" --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Accuracy Test"
-        timeout_in_minutes: 180
-        commands:
-          - export VLLM_HTTP_TIMEOUT_KEEP_ALIVE=120
-          - pytest -s -v tests/e2e/accuracy/test_qwen_image*.py --run-level full_model
-          - pytest -s -v tests/e2e/accuracy/test_diffusers_backend_similarity.py -k '2i_matches_diffusers' -m full_model --run-level full_model
-        mirror_hardwares: h100_1
-
-      - label: ":full_moon: HunyuanImage3-DIT · Accuracy Test"
-        timeout_in_minutes: 180
-        commands:
-          - export HUNYUAN_IMAGE3_MODEL="tencent/HunyuanImage-3.0-Instruct"
-          - export HUNYUAN_IMAGE3_DEVICES="0,1,2,3"
-          - pytest -s -v tests/e2e/accuracy/test_hunyuan_image3_pixel_accuracy.py -m full_model --run-level full_model
-        mirror_hardwares: h100_4
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image"
-        key: nightly-diffusion-x2iat-performance-qwen-image
-        timeout_in_minutes: 180
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
-          - export CACHE_DIT_VERSION=1.3.0
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen_image_vllm_omni.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-        mirror_hardwares: h100_4
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image-Edit"
-        key: nightly-diffusion-x2iat-performance-qwen-image-edit
-        timeout_in_minutes: 180
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
-          - export CACHE_DIT_VERSION=1.3.0
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen_image_edit_vllm_omni.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-        mirror_hardwares: h100_4
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image-Edit-2511"
-        key: nightly-diffusion-x2iat-performance-qwen-image-edit-2511
-        timeout_in_minutes: 180
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
-          - export CACHE_DIT_VERSION=1.3.0
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen_image_edit_2511_vllm_omni.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-        mirror_hardwares: h100_4
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image-Layered"
-        key: nightly-diffusion-x2iat-performance-qwen-image-layered
-        timeout_in_minutes: 120
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
-          - export CACHE_DIT_VERSION=1.3.0
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen_image_layered_vllm_omni.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-        mirror_hardwares: h100_1
-
-      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · BAGEL"
-        key: nightly-diffusion-x2iat-performance-bagel
-        timeout_in_minutes: 180
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
-          - export CACHE_DIT_VERSION=1.3.0
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_bagel_vllm_omni.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-        mirror_hardwares: h100_1
-
-  - group: ":card_index_dividers: Diffusion X2V Model Test"
-    key: nightly-diffusion-x2v-group
-    depends_on: upload-nightly-pipeline
-    if: >-
-      build.env("NIGHTLY") == "1" ||
-      build.pull_request.labels includes "nightly-test" ||
-      build.pull_request.labels includes "diffusion-x2v-test"
-    steps:
-
-      - label: ":full_moon: Diffusion X2V · Wan2.2 T2V Function Test"
-        timeout_in_minutes: 90
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_wan22_expansion.py -k "t2v" -m "full_model and cuda" --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2V · Wan2.2 I2V/TI2V Function Test"
-        timeout_in_minutes: 90
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_wan22_expansion.py -k "i2v" -m "full_model and cuda" --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2V · LingBot Video MoE Function Test"
-        timeout_in_minutes: 90
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_lingbot_video_moe.py -m "full_model and cuda" --run-level "full_model"
-        mirror_hardwares: h100_1
-
-      - label: ":full_moon: Diffusion X2V · Doc Test"
-        timeout_in_minutes: 60
-        commands:
-          - pytest -s -v tests/examples/offline_inference/test_image_to_video.py -m "full_model and example and H100" --run-level "full_model"
-        mirror_hardwares: h100_1
-
-      - label: ":full_moon: Diffusion X2V · Other Function Test"
-        timeout_in_minutes: 90
-        commands:
-          - pytest -s -v tests/e2e/offline_inference/test_wan22_autoround_w4a16_expansion.py -m "full_model and cuda" --run-level "full_model"
-          - pytest -s -v tests/e2e/online_serving/test_wan_2_1_vace_expansion.py tests/e2e/online_serving/test_ltx2_expansion.py -m "full_model and cuda" --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2V · Wan2.2 Accuracy Test"
-        timeout_in_minutes: 180
-        commands:
-          - pytest -s -v tests/e2e/accuracy/wan22_i2v/test_wan22_i2v_video_similarity.py -m full_model --run-level full_model
-          - pytest -s -v tests/e2e/accuracy/test_diffusers_backend_similarity.py -k '2v_matches_diffusers' -m full_model --run-level full_model
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2V · MiniMax H3 T2VA Accuracy Test"
-        timeout_in_minutes: 180
-        commands:
-          - pytest -s -v tests/e2e/accuracy/minimax_h3/test_minimax_h3_t2va_similarity.py -m full_model --run-level full_model
-        mirror_hardwares: h100_4
-
-      - label: ":full_moon: Diffusion X2V · Other Accuracy Test"
-        timeout_in_minutes: 180
-        commands:
-          - pytest -s -v tests/e2e/accuracy/hunyuanvideo15_t2v/test_hunyuanvideo15_t2v_video_similarity.py -m full_model --run-level full_model
-          - pytest -s -v tests/e2e/accuracy/hunyuanvideo15_i2v/test_hunyuanvideo15_i2v_video_similarity.py -m full_model --run-level full_model
-          - pytest -s -v 'tests/e2e/accuracy/ltx/test_ltx_official_similarity.py::test_ltx_one_stage_matches_official[ltx2_3_i2v]' -m full_model --run-level full_model
-          - pytest -s -v tests/e2e/accuracy/test_video_streaming_output_similarity.py -m full_model --run-level full_model
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion X2V · Perf Test"
-        key: nightly-diffusion-x2v-performance
-        timeout_in_minutes: 180
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_wan22_i2v_vllm_omni.json
-            EXIT1=$$?
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_cosmos3_vllm_omni.json
-            EXIT2=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            if [ $$EXIT1 -ne 0 ]; then exit $$EXIT1; fi
-            exit $$EXIT2
-        mirror_hardwares: h100_2
-
-  - group: ":card_index_dividers: Diffusion Test"
-    key: nightly-diffusion-test-group
-    depends_on: upload-nightly-pipeline
-    if: build.env("NIGHTLY") == "1" || build.pull_request.labels includes "nightly-test"
-    steps:
-      - label: ":full_moon: Diffusion · Distributed Test with L4"
-        timeout_in_minutes: 60
-        commands:
-          - pytest -sv tests/diffusion/distributed/ -m 'full_model and cuda and L4' --run-level "full_model"
-        mirror_hardwares: l4_4
-
-      - label: ":full_moon: Diffusion · Distributed Test with H100"
-        timeout_in_minutes: 60
-        commands:
-          - pytest -sv tests/diffusion/distributed/ -m 'full_model and cuda and H100' --run-level "full_model"
-        mirror_hardwares: h100_4
-
-      - label: ":full_moon: Diffusion · Offloader Test"
-        timeout_in_minutes: 60
-        commands:
-          - pytest -sv tests/diffusion/offloader -m "full_model and cuda"
-        mirror_hardwares: l4_1
-
-      - label: ":full_moon: Diffusion · Quantization Test with H100 · Single-GPU"
-        timeout_in_minutes: 60
-        commands:
-          - pytest -sv tests/diffusion/quantization -m 'full_model and cuda and H100' --run-level "full_model"
-        mirror_hardwares: h100_1
-
-      - label: ":full_moon: Diffusion · MiniMax H3 FP8 Quality Test with H100 · 2-GPU"
-        timeout_in_minutes: 60
-        commands:
-          - pytest -sv tests/diffusion/models/minimax_h3/test_minimax_h3_quantization_quality.py -m 'full_model and cuda and H100' --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: ":full_moon: Diffusion · Quantization Test with L4"
-        timeout_in_minutes: 60
-        commands:
-          - pip install "vllm-gguf-plugin>=0.0.3"
-          - pytest -sv tests/diffusion/quantization -m 'full_model and cuda and L4' --run-level "full_model"
-        mirror_hardwares: l4_1
-
-  - label: ":bar_chart: Testcase Statistics"
-    key: nightly-testcase-statistics
-    timeout_in_minutes: 120
-    depends_on: upload-nightly-pipeline
-    if: build.env("NIGHTLY") == "1"
-    commands:
-      - python tools/nightly/buildkite_testcase_statistics.py -o tests/dfx/perf/results/buildkite_testcase_statistics.html
-      - buildkite-agent artifact upload "tests/dfx/perf/results/*.html"
-    mirror_hardwares: h100_1
-
-  # NOTE: ":email: Nightly Collection & Email" is deprecated and will be removed soon.
-  # Do not add new depends_on entries or extend this step.
-  - label: ":email: Nightly Collection & Email"
-    key: nightly-perf-distribution
-    depends_on:
-      - nightly-omni-performance-no-async-chunk
-      - nightly-omni-performance-async-chunk
-      - nightly-omni-performance-vllm-text
-      - nightly-omni-performance-multi-replicas
-      - nightly-tts-performance
-      - nightly-tts-performance-higgs-audio-v3
-      - nightly-diffusion-x2iat-performance-qwen-image
-      - nightly-diffusion-x2iat-performance-qwen-image-edit
-      - nightly-diffusion-x2iat-performance-qwen-image-edit-2511
-      - nightly-diffusion-x2iat-performance-qwen-image-layered
-      - nightly-diffusion-x2iat-performance-bagel
-      - nightly-diffusion-x2v-performance
-      - nightly-testcase-statistics
-    if: build.env("NIGHTLY") == "1" && build.env("EMAIL_DISTRIBUTION") == "1"
-    commands:
-      - pip install openpyxl
-      - export DEFAULT_INPUT_DIR=tests/dfx/perf/results
-      - export DEFAULT_OUTPUT_DIR=tests/dfx/perf/results
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-tts-performance
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-tts-performance-higgs-audio-v3
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-omni-performance-no-async-chunk
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-omni-performance-async-chunk
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-omni-performance-vllm-text
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-omni-performance-multi-replicas
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-qwen-image
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-qwen-image-edit
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-qwen-image-edit-2511
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-qwen-image-layered
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-bagel
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2v-performance
-      - buildkite-agent artifact download "tests/dfx/perf/results/*.html" . --step nightly-testcase-statistics
-      - python tools/nightly/generate_nightly_perf_excel.py
-      - python tools/nightly/generate_nightly_perf_html.py
-      - python tools/nightly/send_nightly_email.py --report-file "tests/dfx/perf/results/*.xlsx, tests/dfx/perf/results/*.html"
-      - buildkite-agent artifact upload "tests/dfx/perf/results/*.xlsx"
-      - buildkite-agent artifact upload "tests/dfx/perf/results/*.html"
-    agents:
-      queue: "cpu_queue_premerge"

.buildkite/cuda/test-ready.yml

@@ -1,346 +0,0 @@
-env:
-  VLLM_WORKER_MULTIPROC_METHOD: spawn
-  VLLM_USE_FLASHINFER_MOE_FP16: 0
-  HF_HUB_DOWNLOAD_TIMEOUT: 300
-  HF_HUB_ETAG_TIMEOUT: 60
-
-steps:
-  - group: ":card_index_dividers: Simple Test"
-    depends_on: upload-ready-pipeline
-    steps:
-      - label: "Simple · Model Executor Test"
-        commands:
-          - "timeout 20m pytest -sv tests/model_executor -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-        mirror_hardwares: l4_1
-
-      - label: "Simple · Diffusion Test"
-        commands:
-          - "timeout 20m pytest -sv tests/diffusion -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-        mirror_hardwares: l4_1
-
-      - label: "Simple · Engine&Entrypoints Test"
-        commands:
-          - "timeout 20m pytest -sv tests/entrypoints tests/engine -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-        mirror_hardwares: l4_1
-
-      - label: "Simple · Other Test"
-        commands:
-          - "timeout 20m pytest -sv tests/ -m 'core_model and cpu' --ignore=tests/diffusion --ignore=tests/model_executor --ignore=tests/entrypoints --ignore=tests/engine --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-        mirror_hardwares: l4_1
-
-  - group: ":card_index_dividers: Diffusion Test"
-    depends_on: upload-ready-pipeline
-    steps:
-
-      - label: "Diffusion · Offloader Test"
-        commands:
-          - timeout 40m pytest -s -v tests/diffusion/offloader -m "core_model and cuda" --run-level "core_model"
-        mirror_hardwares: l4_1
-
-      - label: "Diffusion · Cache Test"
-        commands:
-          - timeout 15m pytest -sv tests/diffusion/cache -m 'core_model and cuda' --run-level "core_model"
-        mirror_hardwares: l4_1
-
-      - label: "Diffusion · Batch Test"
-        commands:
-          - timeout 15m pytest -sv tests/diffusion/batching -m 'core_model and cuda' --run-level "core_model"
-        mirror_hardwares: l4_1
-
-      - label: "Diffusion · Distributed Test"
-        source_file_dependencies:
-          - tests/diffusion/distributed/
-          - vllm_omni/diffusion/distributed/
-          - vllm_omni/diffusion/attention/parallel/
-          - vllm_omni/diffusion/data.py
-          - vllm_omni/diffusion/models/
-        commands:
-          - timeout 20m pytest -sv tests/diffusion/distributed -m 'core_model and cuda' --run-level "core_model"
-        mirror_hardwares: l4_4
-
-      - label: "Diffusion · Model Test"
-        commands:
-          - timeout 15m pytest -sv tests/diffusion/models/ -m 'core_model and cuda' --run-level "core_model"
-        mirror_hardwares: l4_1
-
-      - label: "Diffusion · Other Test"
-        commands:
-          - timeout 15m pytest -sv tests/diffusion/*.py -m 'core_model and cuda' --run-level "core_model"
-        mirror_hardwares: l4_1
-
-  - group: ":card_index_dividers: Tiny Model Tests [Single GPU]"
-    depends_on: upload-ready-pipeline
-    steps:
-
-      - label: "Tiny Model Tests · Diffusion (base)"
-        commands:
-          - timeout 15m pytest -sv tests/model_tests/diffusion -m 'core_model and cuda' --run-level "core_model"
-        mirror_hardwares: l4_1
-
-  - label: "Engine&Model Executor Test"
-    depends_on: upload-ready-pipeline
-    commands:
-      - |
-        timeout 15m bash -c "
-                  pytest -sv tests/engine/ tests/model_executor/ -m 'core_model and cuda' --run-level 'core_model'
-        "
-    mirror_hardwares: l4_1
-
-  - label: "Distributed Test"
-    depends_on: upload-ready-pipeline
-    commands:
-      - |
-        timeout 20m bash -c "
-          pytest -sv tests/distributed/ -m 'core_model and cuda and L4' --run-level 'core_model'
-        "
-    mirror_hardwares: l4_1
-
-  - label: "Entrypoints Test"
-    depends_on: upload-ready-pipeline
-    commands:
-      - |
-        timeout 30m bash -c "
-          export VLLM_WORKER_MULTIPROC_METHOD=spawn
-          pytest -sv tests/entrypoints/ -m 'core_model and cuda' --run-level 'core_model'
-        "
-    mirror_hardwares: l4_1
-
-  - group: ":card_index_dividers: E2E Test"
-    depends_on: upload-ready-pipeline
-    steps:
-      - label: "Custom Pipeline Test"
-        source_file_dependencies:
-          - tests/e2e/features/custom_pipeline/
-          - tests/e2e/features/helpers/custom_pipeline.py
-          - vllm_omni/diffusion/
-          - vllm_omni/engine/
-          - vllm_omni/entrypoints/
-          - vllm_omni/inputs/
-          - vllm_omni/outputs/
-        commands:
-          - timeout 20m pytest -s -v tests/e2e/features/custom_pipeline/ -m "core_model"
-        agents:
-          queue: "gpu_1_queue"
-        plugins:
-          - docker#v5.2.0:
-              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-              always-pull: true
-              propagate-environment: true
-              environment:
-                - "HF_HOME=/fsx/hf_cache"
-              volumes:
-                - "/fsx/hf_cache:/fsx/hf_cache"
-
-      - label: "RLHF · VeRL-Omni E2E Test"
-        source_file_dependencies:
-          - tests/e2e/features/rlhf_test/test_verl_omni_e2e.py
-          - tests/e2e/features/helpers/verl_omni_server.py
-          - tests/e2e/features/helpers/custom_pipeline.py
-          - vllm_omni/diffusion/
-          - vllm_omni/engine/
-          - vllm_omni/entrypoints/
-          - vllm_omni/inputs/
-          - vllm_omni/outputs/
-        commands:
-          - |
-            timeout 20m bash -c "
-              pytest -s -v tests/e2e/features/rlhf_test/test_verl_omni_e2e.py
-            "
-        mirror_hardwares: l4_1
-
-      - label: "Omni · Qwen3-Omni Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen3_omni.py
-          - vllm_omni/model_executor/models/qwen3_omni/
-          - vllm_omni/model_executor/stage_input_processors/qwen3_omni.py
-          - vllm_omni/deploy/qwen3_omni_moe.yaml
-        commands:
-          - |
-            timeout 20m bash -c "
-              pytest -s -v tests/e2e/online_serving/test_qwen3_omni.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_2
-
-      - label: "Omni · MiniCPM-o 4.5 Offline Test"
-        source_file_dependencies:
-          - tests/e2e/offline_inference/test_minicpmo_4_5.py
-          - vllm_omni/deploy/minicpmo_4_5.yaml
-        commands:
-          - |
-            timeout 20m bash -c "
-              pytest -s -v tests/e2e/offline_inference/test_minicpmo_4_5.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_2
-
-      - label: "Omni · MiniCPM-o 4.5 Online Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_minicpmo_4_5.py
-          - vllm_omni/deploy/minicpmo_4_5.yaml
-        commands:
-          - |
-            timeout 20m bash -c "
-              pytest -s -v tests/e2e/online_serving/test_minicpmo_4_5.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_2
-
-      - label: "Omni · MiniCPM-o 4.5 Duplex Test"
-        source_file_dependencies:
-          - tests/assets/minicpmo_4_5/response_required_16k.wav
-          - tests/assets/minicpmo_4_5/soft_interrupt_16k.wav
-          - tests/e2e/online_serving/test_minicpmo_4_5_duplex.py
-          - tests/e2e/online_serving/test_minicpmo_4_5_duplex_expansion.py
-          - tests/e2e/online_serving/test_minicpmo_realtime_duplex_drivers.py
-          - tests/e2e/online_serving/minicpmo_realtime_duplex_scenarios.py
-          - tests/e2e/online_serving/run_minicpmo_realtime_duplex_multi_session.py
-          - tests/e2e/online_serving/run_minicpmo_realtime_duplex_soft_interrupt.py
-          - tests/helpers/minicpmo_4_5_duplex.py
-          - vllm_omni/deploy/minicpmo_4_5_duplex.yaml
-          - vllm_omni/experimental/fullduplex/
-          - vllm_omni/model_executor/models/minicpmo_4_5/
-          - vllm_omni/model_executor/stage_input_processors/minicpmo_4_5_omni.py
-          - vllm_omni/core/sched/
-          - vllm_omni/outputs/multimodal_accumulation.py
-          - vllm_omni/engine/stage_init_utils.py
-          - vllm_omni/engine/async_omni_engine.py
-          - vllm_omni/entrypoints/cli/serve.py
-        commands:
-          - |
-            timeout 20m bash -c "
-              pytest -s -v tests/e2e/online_serving/test_minicpmo_4_5_duplex.py -m 'core_model and cuda' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_2
-
-      - label: "TTS · VoxCPM2 Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_voxcpm2_tts.py
-          - vllm_omni/model_executor/models/voxcpm2/
-          - vllm_omni/deploy/voxcpm2.yaml
-        timeout_in_minutes: 30
-        commands:
-          - |
-            timeout 30m bash -c "
-              export VLLM_LOGGING_LEVEL=DEBUG
-              export VLLM_WORKER_MULTIPROC_METHOD=spawn
-              pytest -s -v tests/e2e/online_serving/test_voxcpm2_tts.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: "TTS · Qwen3-TTS CustomVoice Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen3_tts_customvoice.py
-          - vllm_omni/model_executor/models/qwen3_tts/
-          - vllm_omni/model_executor/stage_input_processors/qwen3_tts.py
-          - vllm_omni/deploy/qwen3_tts.yaml
-        commands:
-          - |
-            timeout 20m bash -c "
-              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
-              pytest -s -v tests/e2e/online_serving/test_qwen3_tts_customvoice.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: "TTS · Qwen3-TTS Base Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen3_tts_base.py
-          - vllm_omni/model_executor/models/qwen3_tts/
-          - vllm_omni/model_executor/stage_input_processors/qwen3_tts.py
-          - vllm_omni/deploy/qwen3_tts.yaml
-        commands:
-          - |
-            timeout 20m bash -c "
-              export VLLM_WORKER_MULTIPROC_METHOD=spawn
-              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
-              export VLLM_OMNI_USE_V2_RUNNER=1
-              pytest -s -v tests/e2e/online_serving/test_qwen3_tts_base.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: "TTS · Higgs-Audio-V3 Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_higgs_audio_v3.py
-          - vllm_omni/model_executor/models/higgs_audio_v3/
-          - vllm_omni/model_executor/stage_input_processors/higgs_audio_v3.py
-          - vllm_omni/deploy/higgs_multimodal_qwen3.yaml
-        commands:
-          - |
-            timeout 20m bash -c "
-              export VLLM_WORKER_MULTIPROC_METHOD=spawn
-              export VLLM_USE_DEEP_GEMM=0
-              export VLLM_MOE_USE_DEEP_GEMM=0
-              pytest -s -v tests/e2e/online_serving/test_higgs_audio_v3.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: l4_1
-
-      - label: "Diffusion · Qwen Image Edit Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen_image_edit.py
-          - vllm_omni/diffusion/models/qwen_image/
-        commands:
-          - |
-            timeout 20m bash -c "
-              pytest -s -v tests/e2e/online_serving/test_qwen_image_edit.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Bagel Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_bagel.py
-          - vllm_omni/model_executor/models/bagel/
-          - vllm_omni/diffusion/models/bagel/
-          - vllm_omni/model_executor/stage_input_processors/bagel.py
-          - vllm_omni/deploy/bagel.yaml
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_IMAGE_FETCH_TIMEOUT=60
-              pytest -s -v tests/e2e/online_serving/test_bagel.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Wan22 Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_wan22_t2v.py
-          - vllm_omni/diffusion/models/wan2_2/
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_IMAGE_FETCH_TIMEOUT=60
-              pytest -s -v tests/e2e/online_serving/test_wan22_t2v.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Cosmos3 Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_cosmos3.py
-          - vllm_omni/diffusion/models/cosmos3/
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_IMAGE_FETCH_TIMEOUT=60
-              pytest -s -v tests/e2e/online_serving/test_cosmos3.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Qwen Image Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen_image.py
-          - vllm_omni/diffusion/models/qwen_image/
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_IMAGE_FETCH_TIMEOUT=60
-              pytest -s -v tests/e2e/online_serving/test_qwen_image.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · Qwen Image Layered Test"
-        source_file_dependencies:
-          - tests/e2e/online_serving/test_qwen_image_layered.py
-          - vllm_omni/diffusion/models/qwen_image/
-        commands:
-          - |
-            timeout 40m bash -c "
-              export VLLM_IMAGE_FETCH_TIMEOUT=60
-              pytest -s -v tests/e2e/online_serving/test_qwen_image_layered.py -m 'core_model' --run-level 'core_model'
-            "
-        mirror_hardwares: h100_1

.buildkite/cuda/test-weekly.yml

@@ -1,136 +0,0 @@
-env:
-  VLLM_WORKER_MULTIPROC_METHOD: spawn
-  VLLM_USE_FLASHINFER_MOE_FP16: 0
-  HF_HUB_DOWNLOAD_TIMEOUT: 300
-  HF_HUB_ETAG_TIMEOUT: 60
-
-steps:
-
-  - label: "Reliability Test - qwen3-omni"
-    timeout_in_minutes: 180
-    depends_on: upload-weekly-pipeline
-    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
-    commands:
-      - pytest -s -v tests/dfx/reliability/test_reliability_qwen3_omni.py -m "slow"
-    mirror_hardwares: h100_2
-
-  - label: "Reliability Test - wan22"
-    timeout_in_minutes: 180
-    depends_on: upload-weekly-pipeline
-    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
-    commands:
-      - pytest -s -v tests/dfx/reliability/test_reliability_wan22.py -m "slow"
-    mirror_hardwares: h100_2
-
-  - label: "Reliability Test - hunyuan-image"
-    timeout_in_minutes: 180
-    depends_on: upload-weekly-pipeline
-    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
-    commands:
-      - pytest -s -v tests/dfx/reliability/test_reliability_hunyuan_image.py -m "slow"
-    mirror_hardwares: h100_4
-
-  - label: "Reliability Test - voxcpm2"
-    timeout_in_minutes: 180
-    depends_on: upload-weekly-pipeline
-    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
-    commands:
-      - pytest -s -v tests/dfx/reliability/test_reliability_voxcpm2.py -m "slow"
-    mirror_hardwares: l4_1
-
-  - group: "Reliability Test - Invalid parameters Test"
-    depends_on: upload-weekly-pipeline
-    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
-    steps:
-
-      - label: "Invalid parameters Test · H100"
-        timeout_in_minutes: 120
-        commands:
-          - pytest -s -v tests/dfx/reliability/invalid_param_test/ -m "slow and H100"
-        mirror_hardwares: h100_2
-
-      - label: "Invalid parameters Test · L4"
-        timeout_in_minutes: 120
-        commands:
-          - pytest -s -v tests/dfx/reliability/invalid_param_test/ -m "slow and L4"
-        mirror_hardwares: l4_1
-
-  - group: ":card_index_dividers: E2E Tests"
-    depends_on: upload-weekly-pipeline
-    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
-    steps:
-
-      - label: "Omni · L4"
-        timeout_in_minutes: 90
-        commands:
-          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-          - pytest -s -v tests/e2e/*/test_qwen2_5_omni_expansion.py --run-level "core_model"
-          - >-
-            pytest -s -v tests/e2e/
-            --ignore=tests/e2e/online_serving/test_qwen2_5_omni_expansion.py
-            --ignore=tests/e2e/offline_inference/test_qwen2_5_omni_expansion.py
-            -m "slow and L4 and omni"
-            --run-level "full_model"
-        mirror_hardwares: l4_4
-
-      - label: "Omni · H100"
-        timeout_in_minutes: 90
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/ -m "slow and H100 and omni"
-            --run-level "full_model"
-        mirror_hardwares: h100_2
-
-      - label: "TTS · L4"
-        timeout_in_minutes: 90
-        commands:
-          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-          - export VLLM_USE_DEEP_GEMM="0"
-          - export VLLM_MOE_USE_DEEP_GEMM="0"
-          - |
-            failed=0
-            for f in $(pytest --collect-only -q tests/e2e/ -m "slow and L4 and tts" --run-level "full_model" | grep '\.py::' | sed -E 's|::.*||' | sort -u); do
-              pytest -s -v "$$f" -m "slow and L4 and tts" --run-level "full_model" || failed=1
-            done
-            exit $$failed
-        mirror_hardwares: l4_1
-
-      - label: "TTS · H100"
-        timeout_in_minutes: 90
-        commands:
-          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-          - export VLLM_USE_DEEP_GEMM="0"
-          - export VLLM_MOE_USE_DEEP_GEMM="0"
-          - >-
-            pytest -sv
-            tests/e2e/ -m "slow and H100 and tts"
-            --run-level "full_model"
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · L4"
-        timeout_in_minutes: 60
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/ -m "slow and diffusion and L4"
-            --run-level "full_model"
-        mirror_hardwares: l4_4
-
-      - label: "Diffusion · H100 · Single-GPU"
-        timeout_in_minutes: 120
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/ -m "slow and diffusion and H100 and not distributed_cuda"
-            --run-level "full_model"
-        mirror_hardwares: h100_1
-
-      - label: "Diffusion · H100 · Multi-GPU"
-        timeout_in_minutes: 90
-        commands:
-          - >-
-            pytest -sv
-            tests/e2e/ -m "slow and diffusion and H100 and distributed_cuda"
-            --run-level "full_model"
-        mirror_hardwares: h100_2

.buildkite/intel/scripts/bootstrap-intel-omni.sh

@@ -1,16 +0,0 @@
-#!/bin/bash
-# vllm-omni Intel bootstrap
-# Uses static pipeline-intel.yml for Intel XPU tests
-
-set -euo pipefail
-
-source .buildkite/common/scripts/resolve_skip_ci.sh
-
-upload_pipeline() {
-    echo "--- 🛠 Preparing Intel pipeline"
-    buildkite-agent pipeline upload .buildkite/intel/pipeline-intel.yml
-}
-
-gate_bootstrap_ci intel l2
-
-upload_pipeline

.buildkite/npu/bootstrap-upload-steps.yml

@@ -1,181 +0,0 @@
-# Uploaded by upload_pipeline.py --upload (not the Buildkite hook).
-# ``if`` is injected by step ``key`` from skip-ci at upload time.
-steps:
-  - label: ":buildkit: Build and Push NPU Test Image - A2"
-    key: image-build-a2
-    agents:
-      queue: "ascend-a2b3"
-      resource_class: "npu-1"
-    plugins:
-      - kubernetes:
-          metadata:
-            annotations:
-              vault.hashicorp.com/agent-init-first: "true"
-              vault.hashicorp.com/agent-inject: "true"
-              vault.hashicorp.com/agent-inject-perms-ca.pem: "0400"
-              vault.hashicorp.com/agent-inject-perms-cert.pem: "0400"
-              vault.hashicorp.com/agent-inject-perms-config.json: "0400"
-              vault.hashicorp.com/agent-inject-perms-key.pem: "0400"
-              vault.hashicorp.com/agent-inject-secret-ca.pem: internal/data/ascend/buildkitd
-              vault.hashicorp.com/agent-inject-secret-cert.pem: internal/data/ascend/buildkitd
-              vault.hashicorp.com/agent-inject-secret-config.json: internal/data/ascend/buildkitd
-              vault.hashicorp.com/agent-inject-secret-key.pem: internal/data/ascend/buildkitd
-              vault.hashicorp.com/agent-inject-template-ca.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.RootCA }}\n{{- end }}"
-              vault.hashicorp.com/agent-inject-template-cert.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.ClientCaCert }}\n{{- end }}"
-              vault.hashicorp.com/agent-inject-template-config.json: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.dockerConfig }}\n{{- end }}"
-              vault.hashicorp.com/agent-inject-template-key.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.ClientCaKey }}\n{{- end }}"
-              vault.hashicorp.com/agent-pre-populate-only: "true"
-              vault.hashicorp.com/agent-run-as-group: "1000"
-              vault.hashicorp.com/agent-run-as-user: "1000"
-              vault.hashicorp.com/agent-service-account-token-volume-name: token-vol
-              vault.hashicorp.com/role: ascend-gha-runners
-              vault.hashicorp.com/secret-volume-path: /home/user/.docker/
-              vault.hashicorp.com/tls-skip-verify: "true"
-          podSpecPatch:
-            volumes:
-            - name: token-vol
-              projected:
-                defaultMode: 420
-                sources:
-                - serviceAccountToken:
-                    audience: api
-                    expirationSeconds: 600
-                    path: token
-    env:
-      VLLM_IMAGE_TAG: "${BUILDKITE_COMMIT}"
-      IMAGE_NAME: "vllm-omni-ci-npu"
-      IMAGE_REGISTRY: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry"
-      BUILDKITD_ADDR: "tcp://buildkitd-service.buildkitd:1234"
-    command: |
-      set -ex
-
-      echo "--- Building and pushing NPU Test Image"
-      echo "Image: $${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG}"
-      echo "buildkitd address: $${BUILDKITD_ADDR}"
-
-      if ! command -v buildctl &> /dev/null; then
-        echo "Installing buildctl..."
-        mkdir -p /tmp/buildkit
-        BUILDKIT_VERSION="v0.29.0"
-        wget -q "https://github.com/moby/buildkit/releases/download/$${BUILDKIT_VERSION}/buildkit-$${BUILDKIT_VERSION}.linux-arm64.tar.gz" -O /tmp/buildkit.tar.gz
-        tar -xzf /tmp/buildkit.tar.gz -C /tmp/buildkit
-        cp /tmp/buildkit/bin/buildctl /usr/local/bin/
-      fi
-
-      export DOCKER_CONFIG=/home/user/.docker
-      buildctl \
-        --addr="$${BUILDKITD_ADDR}" \
-        --tlscacert=/home/user/.docker/ca.pem \
-        --tlscert=/home/user/.docker/cert.pem \
-        --tlskey=/home/user/.docker/key.pem \
-        build \
-        --frontend dockerfile.v0 \
-        --local context=. \
-        --local dockerfile=./docker \
-        --opt filename=Dockerfile.npu.ci \
-        --secret id=dockerconfig,src=/home/user/.docker/config.json \
-        --output type=image,name=$${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG},push=true \
-        --progress=plain
-
-      echo "--- Image pushed successfully"
-      echo "$${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG}"
-
-  - label: ":buildkit: Build and Push NPU Test Image - A3"
-    key: image-build-a3
-    agents:
-      queue: "ascend-a2b3"
-      resource_class: "npu-1"
-    plugins:
-      - kubernetes:
-          metadata:
-            annotations:
-              vault.hashicorp.com/agent-init-first: "true"
-              vault.hashicorp.com/agent-inject: "true"
-              vault.hashicorp.com/agent-inject-perms-ca.pem: "0400"
-              vault.hashicorp.com/agent-inject-perms-cert.pem: "0400"
-              vault.hashicorp.com/agent-inject-perms-config.json: "0400"
-              vault.hashicorp.com/agent-inject-perms-key.pem: "0400"
-              vault.hashicorp.com/agent-inject-secret-ca.pem: internal/data/ascend/buildkitd
-              vault.hashicorp.com/agent-inject-secret-cert.pem: internal/data/ascend/buildkitd
-              vault.hashicorp.com/agent-inject-secret-config.json: internal/data/ascend/buildkitd
-              vault.hashicorp.com/agent-inject-secret-key.pem: internal/data/ascend/buildkitd
-              vault.hashicorp.com/agent-inject-template-ca.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.RootCA }}\n{{- end }}"
-              vault.hashicorp.com/agent-inject-template-cert.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.ClientCaCert }}\n{{- end }}"
-              vault.hashicorp.com/agent-inject-template-config.json: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.dockerConfig }}\n{{- end }}"
-              vault.hashicorp.com/agent-inject-template-key.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.ClientCaKey }}\n{{- end }}"
-              vault.hashicorp.com/agent-pre-populate-only: "true"
-              vault.hashicorp.com/agent-run-as-group: "1000"
-              vault.hashicorp.com/agent-run-as-user: "1000"
-              vault.hashicorp.com/agent-service-account-token-volume-name: token-vol
-              vault.hashicorp.com/role: ascend-gha-runners
-              vault.hashicorp.com/secret-volume-path: /home/user/.docker/
-              vault.hashicorp.com/tls-skip-verify: "true"
-          podSpecPatch:
-            volumes:
-            - name: token-vol
-              projected:
-                defaultMode: 420
-                sources:
-                - serviceAccountToken:
-                    audience: api
-                    expirationSeconds: 600
-                    path: token
-    env:
-      VLLM_IMAGE_TAG: "${BUILDKITE_COMMIT}-a3"
-      IMAGE_NAME: "vllm-omni-ci-npu"
-      IMAGE_REGISTRY: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry"
-      BUILDKITD_ADDR: "tcp://buildkitd-service.buildkitd:1234"
-    command: |
-      set -ex
-
-      echo "--- Building and pushing NPU Test Image"
-      echo "Image: $${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG}"
-      echo "buildkitd address: $${BUILDKITD_ADDR}"
-
-      if ! command -v buildctl &> /dev/null; then
-        echo "Installing buildctl..."
-        mkdir -p /tmp/buildkit
-        BUILDKIT_VERSION="v0.29.0"
-        wget -q "https://github.com/moby/buildkit/releases/download/$${BUILDKIT_VERSION}/buildkit-$${BUILDKIT_VERSION}.linux-arm64.tar.gz" -O /tmp/buildkit.tar.gz
-        tar -xzf /tmp/buildkit.tar.gz -C /tmp/buildkit
-        cp /tmp/buildkit/bin/buildctl /usr/local/bin/
-      fi
-
-      export DOCKER_CONFIG=/home/user/.docker
-      buildctl \
-        --addr="$${BUILDKITD_ADDR}" \
-        --tlscacert=/home/user/.docker/ca.pem \
-        --tlscert=/home/user/.docker/cert.pem \
-        --tlskey=/home/user/.docker/key.pem \
-        build \
-        --frontend dockerfile.v0 \
-        --local context=. \
-        --local dockerfile=./docker \
-        --opt filename=Dockerfile.npu.ci.a3 \
-        --secret id=dockerconfig,src=/home/user/.docker/config.json \
-        --output type=image,name=$${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG},push=true \
-        --progress=plain
-
-      echo "--- Image pushed successfully"
-      echo "$${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG}"
-
-  - label: "Upload Ready Pipeline"
-    depends_on:
-      - image-build-a2
-      - image-build-a3
-    key: upload-ready-pipeline
-    agents:
-      queue: "cpu_queue_premerge"
-    command: |
-      python3 .buildkite/common/scripts/upload_pipeline.py --upload .buildkite/npu/test-npu-ready.yml
-
-  # L4 Test — main+NIGHTLY=1 (scheduled), or PR with specific label (e.g. add label then Rebuild)
-  - label: "Upload Nightly Pipeline"
-    depends_on:
-      - image-build-a2
-      - image-build-a3
-    key: upload-nightly-pipeline
-    commands:
-      - python3 .buildkite/common/scripts/upload_pipeline.py --upload .buildkite/npu/test-npu-nightly.yml
-    agents:
-      queue: "cpu_queue_premerge"

.buildkite/npu/pipeline-npu.yml

@@ -1,9 +0,0 @@
-# Hook uploads this entry YAML; upload_pipeline.py --upload reads bootstrap-upload-steps.yml
-# and injects ``if`` by step ``key`` from skip-ci.
-steps:
-  - label: ":github: Resolve skip-ci & upload pipeline"
-    key: upload-ci-pipeline
-    commands:
-      - "python3 .buildkite/common/scripts/upload_pipeline.py --upload .buildkite/npu/bootstrap-upload-steps.yml"
-    agents:
-      queue: "cpu_queue_premerge"

.buildkite/npu/test-npu-nightly.yml

@@ -1,205 +0,0 @@
-steps:
-  - group: ":card_index_dividers: Omni Model Test"
-    key: nightly-omni-test-group
-    depends_on: upload-nightly-pipeline
-    if: >-
-      build.env("NIGHTLY") == "1" ||
-      build.pull_request.labels includes "nightly-test" ||
-      build.pull_request.labels includes "omni-test"
-    steps:
-      - label: ":full_moon: Omni · Perf Test · No Async Chunk"
-        key: nightly-omni-performance-no-async-chunk
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_4
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py -m npu --test-config-file tests/dfx/perf/tests/test_qwen3_omni_no_async_chunk.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-      - label: ":full_moon: Omni · Perf Test · Async Chunk"
-        key: nightly-omni-performance-async-chunk
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_4
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py -m npu --test-config-file tests/dfx/perf/tests/test_qwen3_omni_async_chunk.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-      - label: ":robot: Omni · MiniCPM-o 4.5 Test"
-        mirror_hardwares: a3_npu_2
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-        commands:
-          - pytest -s -v tests/e2e/offline_inference/test_minicpmo_4_5.py tests/e2e/online_serving/test_minicpmo_4_5.py tests/e2e/online_serving/test_minicpmo_4_5_expansion.py -m 'full_model' --run-level 'full_model'
-  - group: ":card_index_dividers: TTS Model Test"
-    key: nightly-tts-test-group
-    depends_on: upload-nightly-pipeline
-    if: build.env("NIGHTLY") == "1" || build.pull_request.labels includes "nightly-test" || build.pull_request.labels includes "tts-test"
-    steps:
-      - label: ":full_moon: TTS · Perf Test"
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_2
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py -m npu --test-config-file tests/dfx/perf/tests/test_tts.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
-            exit $$EXIT
-  - group: ":card_index_dividers: Diffusion X2I(&A&T) Model Test"
-    key: nightly-diffusion-x2iat-group
-    depends_on: upload-nightly-pipeline
-    if: >-
-      build.env("NIGHTLY") == "1" ||
-      build.pull_request.labels includes "nightly-test" ||
-      build.pull_request.labels includes "diffusion-x2iat-test"
-    steps:
-      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image-Edit-2511"
-        key: nightly-diffusion-x2iat-performance-qwen-image-edit-2511
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_8
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-          DIFFUSION_ATTENTION_BACKEND: TORCH_SDPA
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - export CACHE_DIT_VERSION=1.3.0
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py -m npu --test-config-file tests/dfx/perf/tests/test_qwen_image_edit_2511_vllm_omni.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-  - group: ":card_index_dividers: Diffusion I2I Model Test"
-    key: nightly-HunyuanImage3-x2i-test-group
-    depends_on: upload-nightly-pipeline
-    if: >-
-      build.env("NIGHTLY") == "1" ||
-      build.pull_request.labels includes "nightly-test" ||
-      build.pull_request.labels includes "diffusion-x2iat-test"
-    steps:
-      - label: ":full_moon: HunyuanImage3-DIT · Accuracy Test"
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_4
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          HUNYUAN_IMAGE3_MODEL: "tencent/HunyuanImage-3.0-Instruct"
-          HUNYUAN_IMAGE3_DEVICES: "0,1,2,3"
-          DIFFUSION_ATTENTION_BACKEND: TORCH_SDPA
-        commands:
-          - pytest -s -v tests/e2e/accuracy/test_hunyuan_image3_pixel_accuracy.py -m "full_model and A3" --run-level full_model
-  - group: ":card_index_dividers: Diffusion Benchmark Perf Test"
-    key: nightly-diffusion-benchmark-group
-    depends_on: upload-nightly-pipeline
-    if: >-
-      build.env("NIGHTLY") == "1" ||
-      build.pull_request.labels includes "nightly-test" ||
-      build.pull_request.labels includes "diffusion-x2iat-test"
-    steps:
-      - label: ":full_moon: Diffusion Perf · HunyuanImage3 TP=2 CFGP=2"
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_4
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          HF_TOKEN: "${HF_TOKEN}"
-          DIFFUSION_ATTENTION_BACKEND: TORCH_SDPA
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py -m A3 --test-config-file tests/dfx/perf/tests/test_hunyuan_image_tp2_cfgp2.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-      - label: ":full_moon: Diffusion Perf · HunyuanImage3 TP=2 SP=2"
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_4
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          HF_TOKEN: "${HF_TOKEN}"
-          DIFFUSION_ATTENTION_BACKEND: TORCH_SDPA
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py -m A3 --test-config-file tests/dfx/perf/tests/test_hunyuan_image_tp2_sp2.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-      - label: ":full_moon: Diffusion Perf · HunyuanImage3 TP=4"
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_4
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          HF_TOKEN: "${HF_TOKEN}"
-          DIFFUSION_ATTENTION_BACKEND: TORCH_SDPA
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py -m A3 --test-config-file tests/dfx/perf/tests/test_hunyuan_image_tp4.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT
-  - group: ":card_index_dividers: Diffusion X2V Model Test"
-    key: nightly-diffusion-x2v-group
-    depends_on: upload-nightly-pipeline
-    if: >-
-      build.env("NIGHTLY") == "1" ||
-      build.pull_request.labels includes "nightly-test" ||
-      build.pull_request.labels includes "diffusion-x2v-test"
-    steps:
-      - label: ":full_moon: Diffusion X2V · Function Test"
-        mirror_hardwares: a2b3_npu_4
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_wan22_expansion.py -k npu_i2v --run-level "full_model"
-      - label: ":full_moon: Diffusion X2V · Perf Test"
-        key: nightly-diffusion-x2v-performance
-        timeout_in_minutes: 180
-        mirror_hardwares: a3_npu_4
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-          DIFFUSION_ATTENTION_BACKEND: TORCH_SDPA
-        commands:
-          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
-          - |
-            set +e
-            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py -m npu --test-config-file tests/dfx/perf/tests/test_wan22_i2v_vllm_omni.json
-            EXIT=$$?
-            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
-            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
-            exit $$EXIT

.buildkite/npu/test-npu-ready.yml

@@ -1,34 +0,0 @@
-steps:
-  - group: ":card_index_dividers: Omni Model Test"
-    key: ready-omni-group
-    depends_on: upload-ready-pipeline
-    steps:
-      - label: ":robot: Omni · MiniCPM-o 4.5 Test"
-        agents:
-          queue: "ascend-a3"
-          resource_class: "npu-2"
-        image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}-a3"
-        plugins:
-          - kubernetes:
-              podSpecPatch:
-                imagePullSecrets:
-                  - name: swr-secret
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-        commands:
-          - pytest -s -v tests/e2e/offline_inference/test_minicpmo_4_5.py tests/e2e/online_serving/test_minicpmo_4_5.py -m 'core_model' --run-level 'core_model'
-
-  - group: ":card_index_dividers: TTS Model Test"
-    key: ready-tts-group
-    depends_on: upload-ready-pipeline
-    steps:
-      - label: ":sound: TTS · Qwen3-TTS CustomVoice Test"
-        mirror_hardwares: a2b3_npu_1
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-          HF_TOKEN: "${HF_TOKEN}"
-        commands:
-          - pytest -s -v tests/e2e/online_serving/test_qwen3_tts_customvoice.py -m 'core_model and npu' --run-level 'core_model'

.buildkite/pipeline-intel.yaml

@@ -1,16 +1,16 @@
-# .buildkite/intel/pipeline-intel.yml
+# .buildkite/pipeline-intel.yaml
 steps:
   - group: "INTEL Tests"
     steps:
       - label: "Intel: XPU Test"
-        command: "bash .buildkite/intel/scripts/run-xpu-test.sh"
+        command: "bash .buildkite/scripts/hardware_ci/run-xpu-test.sh"
         agents:
           queue: "intel-gpu-omni"
         env:
           DOCKER_BUILDKIT: "1"
           # Buildkite will automatically replace this with the actual commit hash
           VLLM_IMAGE_TAG: "${BUILDKITE_COMMIT}"
-          VLLM_VERSION: "v0.26.0"
+          VLLM_VERSION: "v0.23.0"
         priority: 100
         timeout_in_minutes: 60
         soft_fail: false

.buildkite/pipeline-npu-a3.yaml

@@ -88,7 +88,7 @@
 #   #     queue: "ascend-a3"
 #   #     resource_class: "npu-2"
 #   #   command: |
-#   #     buildkite-agent pipeline upload .buildkite/npu/test-npu-ready.yml
+#   #     buildkite-agent pipeline upload .buildkite/test-npu-ready.yaml
 
   # # L4 Test — main+NIGHTLY=1 (scheduled), or PR with specific label (e.g. add label then Rebuild)
   # - label: "Upload Nightly Pipeline"
@@ -104,7 +104,7 @@
   #       build.pull_request.labels includes "diffusion-x2v-test"
   #     ))
   #   commands:
-  #     - buildkite-agent pipeline upload .buildkite/npu/test-npu-nightly.yml
+  #     - buildkite-agent pipeline upload .buildkite/test-npu-nightly.yaml
   #   agents:
   #     queue: "ascend-a3"
   #     resource_class: "npu-2"

.buildkite/pipeline-npu.yaml

@@ -0,0 +1,112 @@
+steps:
+  - label: ":buildkit: Build and Push NPU Test Image - A2"
+    key: image-build
+    agents:
+      queue: "ascend-a2b3"
+      resource_class: "npu-1"
+    plugins:
+      - kubernetes:
+          metadata:
+            annotations:
+              vault.hashicorp.com/agent-init-first: "true"
+              vault.hashicorp.com/agent-inject: "true"
+              vault.hashicorp.com/agent-inject-perms-ca.pem: "0400"
+              vault.hashicorp.com/agent-inject-perms-cert.pem: "0400"
+              vault.hashicorp.com/agent-inject-perms-config.json: "0400"
+              vault.hashicorp.com/agent-inject-perms-key.pem: "0400"
+              vault.hashicorp.com/agent-inject-secret-ca.pem: internal/data/ascend/buildkitd
+              vault.hashicorp.com/agent-inject-secret-cert.pem: internal/data/ascend/buildkitd
+              vault.hashicorp.com/agent-inject-secret-config.json: internal/data/ascend/buildkitd
+              vault.hashicorp.com/agent-inject-secret-key.pem: internal/data/ascend/buildkitd
+              vault.hashicorp.com/agent-inject-template-ca.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.RootCA }}\n{{- end }}"
+              vault.hashicorp.com/agent-inject-template-cert.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.ClientCaCert }}\n{{- end }}"
+              vault.hashicorp.com/agent-inject-template-config.json: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.dockerConfig }}\n{{- end }}"
+              vault.hashicorp.com/agent-inject-template-key.pem: "{{- with secret \"internal/data/ascend/buildkitd\" -}}\n{{ .Data.data.ClientCaKey }}\n{{- end }}"
+              vault.hashicorp.com/agent-pre-populate-only: "true"
+              vault.hashicorp.com/agent-run-as-group: "1000"
+              vault.hashicorp.com/agent-run-as-user: "1000"
+              vault.hashicorp.com/agent-service-account-token-volume-name: token-vol
+              vault.hashicorp.com/role: ascend-gha-runners
+              vault.hashicorp.com/secret-volume-path: /home/user/.docker/
+              vault.hashicorp.com/tls-skip-verify: "true"
+          podSpecPatch:
+            volumes:
+            - name: token-vol
+              projected:
+                defaultMode: 420
+                sources:
+                - serviceAccountToken:
+                    audience: api
+                    expirationSeconds: 600
+                    path: token
+    env:
+      VLLM_IMAGE_TAG: "${BUILDKITE_COMMIT}"
+      IMAGE_NAME: "vllm-omni-ci-npu"
+      IMAGE_REGISTRY: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry"
+      BUILDKITD_ADDR: "tcp://buildkitd-service.buildkitd:1234"
+    command: |
+      set -ex
+
+      echo "--- Building and pushing NPU Test Image"
+      echo "Image: $${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG}"
+      echo "buildkitd address: $${BUILDKITD_ADDR}"
+
+      if ! command -v buildctl &> /dev/null; then
+        echo "Installing buildctl..."
+        mkdir -p /tmp/buildkit
+        BUILDKIT_VERSION="v0.29.0"
+        wget -q "https://github.com/moby/buildkit/releases/download/$${BUILDKIT_VERSION}/buildkit-$${BUILDKIT_VERSION}.linux-arm64.tar.gz" -O /tmp/buildkit.tar.gz
+        tar -xzf /tmp/buildkit.tar.gz -C /tmp/buildkit
+        cp /tmp/buildkit/bin/buildctl /usr/local/bin/
+      fi
+
+      export DOCKER_CONFIG=/home/user/.docker
+      buildctl \
+        --addr="$${BUILDKITD_ADDR}" \
+        --tlscacert=/home/user/.docker/ca.pem \
+        --tlscert=/home/user/.docker/cert.pem \
+        --tlskey=/home/user/.docker/key.pem \
+        build \
+        --frontend dockerfile.v0 \
+        --local context=. \
+        --local dockerfile=./docker \
+        --opt filename=Dockerfile.npu.ci \
+        --secret id=dockerconfig,src=/home/user/.docker/config.json \
+        --output type=image,name=$${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG},push=true \
+        --progress=plain
+
+      echo "--- Image pushed successfully"
+      echo "$${IMAGE_REGISTRY}/$${IMAGE_NAME}:$${VLLM_IMAGE_TAG}"
+
+  - label: "Upload Ready Pipeline"
+    depends_on: image-build
+    key: upload-ready-pipeline
+    if: >-
+      build.branch != "main" && (
+        build.pull_request.labels includes "npu-test" ||
+        build.pull_request.labels includes "ready"
+      )
+    agents:
+      queue: "ascend-a2b3"
+      resource_class: "npu-1"
+    command: |
+      buildkite-agent pipeline upload .buildkite/test-npu-ready.yaml
+
+  # L4 Test — main+NIGHTLY=1 (scheduled), or PR with specific label (e.g. add label then Rebuild)
+  - label: "Upload Nightly Pipeline"
+    depends_on: image-build
+    key: upload-nightly-pipeline
+    if: >-
+      (build.branch == "main" && build.env("NIGHTLY") == "1") ||
+      (build.branch != "main" && (
+        build.pull_request.labels includes "nightly-test" ||
+        build.pull_request.labels includes "omni-test" ||
+        build.pull_request.labels includes "tts-test" ||
+        build.pull_request.labels includes "diffusion-x2iat-test" ||
+        build.pull_request.labels includes "diffusion-x2v-test"
+      ))
+    commands:
+      - buildkite-agent pipeline upload .buildkite/test-npu-nightly.yaml
+    agents:
+      queue: "ascend-a2b3"
+      resource_class: "npu-1"

.buildkite/pipeline.yml

@@ -0,0 +1,77 @@
+# Document 1: Buildkite loads only this block on first parse. The next step resolves docs-only,
+# pytest skip-mark-only, or combined skip-ci from git diff, then uploads document 2. When skip-ci
+# applies, image-build still runs for the nightly exception in upload_pipeline.py (nightly-test
+# label or main NIGHTLY=1); otherwise upload-nightly can be skipped together with test-ready/test-merge.
+# Weekly uploads also depend on image-build.
+
+# Document 2: appended after `---`; same file, read by upload_pipeline.py (not evaluated as a second pipeline by Buildkite). Child uploads include test-ready.yml, test-merge.yml, test-nightly.yml, and test-weekly.yml.
+steps:
+  - label: ":github: Resolve skip-ci (docs / skip marks) & upload pipeline"
+    key: upload-ci-pipeline
+    commands:
+      - "python3 .buildkite/scripts/upload_pipeline.py --upload .buildkite/pipeline.yml"
+    agents:
+      queue: "cpu_queue_premerge"
+
+---
+steps:
+  - label: ":docker: Build image"
+    key: image-build
+    if: __IMAGE_BUILD_IF__
+    commands:
+      - "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7"
+      - "docker build --progress=plain --file docker/Dockerfile.ci -t vllm-omni-ci ."
+      - "docker tag vllm-omni-ci public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT"
+      - "docker push public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT"
+    agents:
+      queue: "cpu_queue_premerge"
+
+  # L2 Test
+  - label: "Upload Ready Pipeline"
+    depends_on: image-build
+    key: upload-ready-pipeline
+    if: __UPLOAD_READY_IF__
+    commands:
+      - python3 .buildkite/scripts/upload_pipeline.py --upload .buildkite/test-ready.yml
+    agents:
+      queue: "cpu_queue_premerge"
+
+  # L3 Test
+  - label: "Upload Merge Pipeline"
+    depends_on: image-build
+    key: upload-merge-pipeline
+    if: __UPLOAD_MERGE_IF__
+    commands:
+      - python3 .buildkite/scripts/upload_pipeline.py --upload .buildkite/test-merge.yml
+    agents:
+      queue: "cpu_queue_premerge"
+
+  # L4 Test — main+NIGHTLY=1 (scheduled), or PR with specific label (e.g. add label then Rebuild)
+  - label: "Upload Nightly Pipeline"
+    depends_on: image-build
+    key: upload-nightly-pipeline
+    if: >-
+      (build.branch == "main" && build.env("NIGHTLY") == "1") ||
+      (build.branch != "main" && (
+        build.pull_request.labels includes "nightly-test" ||
+        build.pull_request.labels includes "omni-test" ||
+        build.pull_request.labels includes "tts-test" ||
+        build.pull_request.labels includes "diffusion-x2iat-test" ||
+        build.pull_request.labels includes "diffusion-x2v-test"
+      ))
+    commands:
+      - buildkite-agent pipeline upload .buildkite/test-nightly.yml
+    agents:
+      queue: "cpu_queue_premerge"
+
+  # L5 Test — main+WEEKLY=1 (scheduled), or PR with weekly-test label
+  - label: "Upload Weekly Pipeline"
+    depends_on: image-build
+    key: upload-weekly-pipeline
+    if: >-
+      (build.branch == "main" && build.env("WEEKLY") == "1") ||
+      (build.branch != "main" && build.pull_request.labels includes "weekly-test")
+    commands:
+      - buildkite-agent pipeline upload .buildkite/test-weekly.yml
+    agents:
+      queue: "cpu_queue_premerge"

.buildkite/rebase-pipeline.yaml

@@ -2,7 +2,10 @@ steps:
   - label: ":docker: Build image"
     key: image-build
     commands:
-      - "bash .buildkite/cuda/scripts/build-ci-image.sh"
+      - "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7"
+      - "docker build --progress=plain --file docker/Dockerfile.ci -t vllm-omni-ci ."
+      - "docker tag vllm-omni-ci public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT"
+      - "docker push public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT"
     agents:
       queue: "cpu_queue_premerge"
 
@@ -11,7 +14,7 @@ steps:
     depends_on: image-build
     key: upload-ready-pipeline
     commands:
-      - python3 .buildkite/common/scripts/upload_pipeline.py --upload --all .buildkite/cuda/test-ready.yml
+      - buildkite-agent pipeline upload .buildkite/test-ready.yml
     agents:
       queue: "cpu_queue_premerge"
 
@@ -20,7 +23,7 @@ steps:
     depends_on: image-build
     key: upload-merge-pipeline
     commands:
-      - python3 .buildkite/common/scripts/upload_pipeline.py --upload --all .buildkite/cuda/test-merge.yml
+      - buildkite-agent pipeline upload .buildkite/test-merge.yml
     agents:
       queue: "cpu_queue_premerge"
 
@@ -30,6 +33,6 @@ steps:
     key: upload-nightly-pipeline
     if: build.env("NIGHTLY") == "1"
     commands:
-      - python3 .buildkite/common/scripts/upload_pipeline.py --upload --all .buildkite/cuda/test-nightly.yml
+      - buildkite-agent pipeline upload .buildkite/test-nightly.yml
     agents:
       queue: "cpu_queue_premerge"

.buildkite/release-pipeline.yaml

@@ -16,7 +16,7 @@ steps:
       - "uv venv --python=3.12 && source .venv/bin/activate"
       - "uv pip install --upgrade build"
       - "python3 -m build"
-      - "bash .buildkite/release/scripts/upload-nightly-wheels.sh"
+      - "bash .buildkite/scripts/upload-nightly-wheels.sh"
 
   - label: "Generate and upload wheel indices"
     depends_on: "build-wheel"
@@ -24,7 +24,7 @@ steps:
     agents:
       queue: small_cpu_queue_release
     commands:
-      - "bash .buildkite/release/scripts/generate-and-upload-nightly-index.sh"
+      - "bash .buildkite/scripts/generate-and-upload-nightly-index.sh"
 
   - block: "Upload release wheels to PyPI"
     depends_on: "build-wheel"
@@ -38,7 +38,7 @@ steps:
     agents:
       queue: small_cpu_queue_release
     commands:
-      - "bash .buildkite/release/scripts/publish-release-wheels-pypi.sh"
+      - "bash .buildkite/scripts/publish-release-wheels-pypi.sh"
 
   - block: "Unblock to build release Docker images"
     depends_on: ~
@@ -100,7 +100,7 @@ steps:
     agents:
       queue: small_cpu_queue_release
     commands:
-      - "bash .buildkite/release/scripts/publish-release-images.sh"
+      - "bash .buildkite/scripts/publish-release-images.sh"
     plugins:
       - docker-login#v3.0.0:
           username: vllmbot

.buildkite/scripts/docker_login_ecr_public.sh

@@ -0,0 +1,73 @@
+#!/bin/bash
+# Helper function to safely login to ECR Public with per-job config isolation
+# Uses DOCKER_CONFIG environment variable to prevent race conditions
+#
+# This script prevents the "device or resource busy" error by giving each
+# Buildkite job its own isolated Docker config directory.
+#
+# Usage:
+#   source docker_login_ecr_public.sh && safe_docker_login_ecr_public
+
+set -euo pipefail
+
+# Configuration
+ECR_REGISTRY="public.ecr.aws"
+
+setup_isolated_docker_config() {
+    # Use BUILDKITE_JOB_ID for job-specific isolation
+    # Fallback to PID if running outside Buildkite
+    local job_id="${BUILDKITE_JOB_ID:-$$}"
+
+    # Set Docker config to job-specific directory
+    export DOCKER_CONFIG="/tmp/docker-config-${job_id}"
+
+    # Create directory if it doesn't exist
+    mkdir -p "$DOCKER_CONFIG"
+
+    echo "[docker-config] Using isolated Docker config: $DOCKER_CONFIG"
+}
+
+check_docker_auth() {
+    # Check if already authenticated to the given registry
+    # Returns 0 if authenticated, 1 if not
+    local registry="$1"
+
+    # Check if credentials exist in the isolated config
+    if [[ -f "$DOCKER_CONFIG/config.json" ]]; then
+        # Check if registry is present in config
+        if grep -q "$registry" "$DOCKER_CONFIG/config.json" 2>/dev/null; then
+            return 0
+        fi
+    fi
+
+    return 1
+}
+
+safe_docker_login_ecr_public() {
+    # Setup isolated config first
+    setup_isolated_docker_config
+
+    local registry="$ECR_REGISTRY"
+
+    # Check if already authenticated (within this job)
+    if check_docker_auth "$registry"; then
+        echo "[docker-login] Already authenticated to $registry in this job"
+        return 0
+    fi
+
+    # Perform login to isolated config directory
+    echo "[docker-login] Logging in to $ECR_REGISTRY (isolated config)..."
+    if aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin "$ECR_REGISTRY"; then
+        echo "[docker-login] Login successful (config: $DOCKER_CONFIG)"
+        return 0
+    else
+        local exit_code=$?
+        echo "[docker-login] ERROR: Login failed with exit code $exit_code" >&2
+        return $exit_code
+    fi
+}
+
+# Execute if run as script (not sourced)
+if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then
+    safe_docker_login_ecr_public
+fi

.buildkite/scripts/generate-and-upload-nightly-index.sh

@@ -36,12 +36,12 @@ mkdir -p "$INDICES_OUTPUT_DIR"
 
 # HACK: we do not need regex module here, but it is required by pre-commit hook
 # To avoid any external dependency, we simply replace it back to the stdlib re module
-sed -i.bak 's/import regex as re/import re/g' .buildkite/release/scripts/generate-nightly-index.py && rm -f .buildkite/release/scripts/generate-nightly-index.py.bak
+sed -i.bak 's/import regex as re/import re/g' .buildkite/scripts/generate-nightly-index.py && rm -f .buildkite/scripts/generate-nightly-index.py.bak
 
 # Generate indices -- the version is just the commit hash (not omni/{commit})
 # because relative paths are computed between the index and wheel directories,
 # both of which live under the omni/ prefix in S3.
-$PYTHON .buildkite/release/scripts/generate-nightly-index.py \
+$PYTHON .buildkite/scripts/generate-nightly-index.py \
     --version "$BUILDKITE_COMMIT" \
     --current-objects "$obj_json" \
     --output-dir "$INDICES_OUTPUT_DIR" \
@@ -78,7 +78,7 @@ if [[ "$version" != *"dev"* ]]; then
     rm -rf "${INDICES_OUTPUT_DIR:?}"
     mkdir -p "$INDICES_OUTPUT_DIR"
     # wheel-dir is overridden to be the commit directory, so that the indices point to the correct wheel path
-    $PYTHON .buildkite/release/scripts/generate-nightly-index.py \
+    $PYTHON .buildkite/scripts/generate-nightly-index.py \
         --version "$s3_version" \
         --wheel-dir "$BUILDKITE_COMMIT" \
         --current-objects "$obj_json" \

.buildkite/scripts/hardware_ci/run-amd-test.sh

@@ -56,7 +56,7 @@ container_name="rocm_${BUILDKITE_COMMIT}_$(tr -dc A-Za-z0-9 < /dev/urandom | hea
 # # Install AWS CLI to authenticate to ECR Public Gallery to get higher rate limit for pulling images
 # sudo apt-get update && sudo apt-get install -y awscli
 ##  Use safe docker login helper to prevent race conditions
-# source "$(dirname "${BASH_SOURCE[0]}")/../../common/scripts/docker_login_ecr_public.sh"
+# source "$(dirname "${BASH_SOURCE[0]}")/../docker_login_ecr_public.sh"
 # safe_docker_login_ecr_public
 
 ## Pull the container from AMD Docker Hub

.buildkite/scripts/hardware_ci/run-xpu-test.sh

@@ -13,33 +13,12 @@ container_name="xpu_${BUILDKITE_COMMIT}_$(
 )"
 
 cd "${omni_source_dir}"
-
-# The XPU base image is ~37GB; the default gzip layer exporter is single-threaded
-# and takes ~16min to compress it. zstd is multi-threaded (uses all cores) at a
-# similar ratio; set EXPORT_COMPRESSION=uncompressed to skip compression entirely
-# for local-only images. This requires the containerd image store (buildx docker
-# driver), which is the default here.
-EXPORT_COMPRESSION="${EXPORT_COMPRESSION:-zstd}"
-if [ "${EXPORT_COMPRESSION}" = "uncompressed" ]; then
-    export_args=(--output "type=image,name={{IMAGE}},compression=uncompressed")
-else
-    export_args=(--output "type=image,name={{IMAGE}},compression=${EXPORT_COMPRESSION},compression-level=3,force-compression=true")
-fi
-
-docker_build() {
-    # $1 = image name; remaining args passed through to docker build.
-    local image="$1"
-    shift
-    local out=("${export_args[@]/'{{IMAGE}}'/${image}}")
-    docker build "${out[@]}" "$@" -f docker/Dockerfile.xpu .
-}
-
 if [ -z "$(docker images -q "${base_image_name}")" ]; then
-    docker_build "${base_image_name}" --target vllm-base --build-arg "VLLM_VERSION=${VLLM_VERSION}"
+    docker build --target vllm-base -t "${base_image_name}" --build-arg "VLLM_VERSION=${VLLM_VERSION}" -f docker/Dockerfile.xpu .
 fi
 
 # Try building the docker image
-docker_build "${image_name}" --build-arg "VLLM_BASE=${base_image_name}" --build-arg "VLLM_VERSION=${VLLM_VERSION}"
+docker build --build-arg "VLLM_BASE=${base_image_name}" --build-arg "VLLM_VERSION=${VLLM_VERSION}" -t "${image_name}" -f docker/Dockerfile.xpu .
 
 # Setup cleanup
 remove_docker_container() {
@@ -74,6 +53,6 @@ time timeout -k 30 30m docker run \
     cd /workspace/vllm-omni
     pytest -v -s -m "core_model and xpu and B60"
     pytest -v -s tests/diffusion/quantization/test_mxfp8_config.py
+    export VLLM_XPU_USE_SAMPLER_KERNEL=0    # NOTE: Remove this after vLLM v0.21.1 is merged. Fixes Qwen2-5 omni-expansion tests.
     pytest -v -s -m "advanced_model and xpu and B60"
-    pytest -v -s -m "omni and xpu and B60"
 '

.buildkite/scripts/hardware_ci/run_npu_test.sh

@@ -0,0 +1,145 @@
+#!/bin/bash
+
+# This script build the Ascend NPU docker image and run the offline inference inside the container.
+# It serves a sanity check for compilation and basic model usage.
+set -ex
+
+# Base ubuntu image with basic ascend development libraries and python installed
+VLLM_OMNI_REPO="https://github.com/vllm-project/vllm-omni.git"
+BASE_IMAGE_NAME="quay.nju.edu.cn/ascend/vllm-ascend:v0.11.0rc2"
+image_name="npu/vllm-omni-ci:${BUILDKITE_COMMIT}_${EPOCHSECONDS}"
+# image_name="npu/vllm-ci:${BUILDKITE_COMMIT}_${EPOCHSECONDS}"
+container_name="npu_${BUILDKITE_COMMIT}_$(tr -dc A-Za-z0-9 < /dev/urandom | head -c 10; echo)"
+
+# BUILDKITE_AGENT_NAME format is {hostname}-{agent_idx}-{npu_card_num}cards
+agent_idx=$(echo "${BUILDKITE_AGENT_NAME}" | awk -F'-' '{print $(NF-1)}')
+echo "agent_idx: ${agent_idx}"
+builder_name="cachebuilder${agent_idx}"
+builder_cache_dir="/mnt/docker-cache${agent_idx}"
+mkdir -p ${builder_cache_dir}
+
+# Try building the docker image
+cat <<EOF | DOCKER_BUILDKIT=1 docker build \
+    --add-host pypi-cache:${PYPI_CACHE_HOST} \
+    --builder ${builder_name} --cache-from type=local,src=${builder_cache_dir} \
+                           --cache-to type=local,dest=${builder_cache_dir},mode=max \
+    --build-arg BUILDKITE_PULL_REQUEST="${BUILDKITE_PULL_REQUEST}" \
+    --build-arg BUILDKITE_PULL_REQUEST_REPO="${BUILDKITE_PULL_REQUEST_REPO}" \
+    --progress=plain --load -t ${image_name} -f - .
+FROM ${BASE_IMAGE_NAME}
+
+# Define environments
+ENV DEBIAN_FRONTEND=noninteractive
+
+RUN pip config set global.index-url http://pypi-cache:${PYPI_CACHE_PORT}/pypi/simple && \
+    pip config set global.trusted-host pypi-cache && \
+    apt-get update -y && \
+    apt-get install -y python3-pip git vim wget net-tools gcc g++ cmake libnuma-dev && \
+    rm -rf /var/cache/apt/* && \
+    rm -rf /var/lib/apt/lists/*
+
+# Install for pytest to make the docker build cache layer always valid
+RUN --mount=type=cache,target=/root/.cache/pip \
+    pip install pytest>=6.0  pytest-cov modelscope
+
+COPY . .
+
+# Install vllm-omni
+WORKDIR /workspace
+ARG VLLM_OMNI_REPO=https://github.com/vllm-project/vllm-omni.git
+ARG VLLM_OMNI_TAG=main
+ARG BUILDKITE_PULL_REQUEST
+ARG BUILDKITE_PULL_REQUEST_REPO
+RUN git config --global url."https://gh-proxy.test.osinfra.cn/https://github.com/".insteadOf "https://github.com/" && \
+    if [ "\$BUILDKITE_PULL_REQUEST" != "false" ] && [ -n "\$BUILDKITE_PULL_REQUEST" ]; then \
+        echo "Cloning and checking out PR #\$BUILDKITE_PULL_REQUEST..." && \
+        git clone \$VLLM_OMNI_REPO /workspace/vllm-omni && \
+        cd /workspace/vllm-omni && \
+        git fetch origin pull/\$BUILDKITE_PULL_REQUEST/head:pr-\$BUILDKITE_PULL_REQUEST && \
+        git checkout pr-\$BUILDKITE_PULL_REQUEST; \
+    else \
+        echo "Not a PR build, using main branch" && \
+        git clone --depth 1 \$VLLM_OMNI_REPO /workspace/vllm-omni; \
+    fi
+
+RUN --mount=type=cache,target=/root/.cache/pip \
+    export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi && \
+    source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
+    source /usr/local/Ascend/nnal/atb/set_env.sh && \
+    export LD_LIBRARY_PATH=\$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
+    python3 -m pip install -v -e /workspace/vllm-omni/
+
+ENV VLLM_WORKER_MULTIPROC_METHOD=spawn
+
+WORKDIR /workspace/vllm-omni
+CMD ["/bin/bash"]
+
+EOF
+
+# Setup cleanup
+remove_docker_container() {
+  docker rm -f "${container_name}" || true;
+  docker image rm -f "${image_name}" || true;
+  docker system prune -f || true;
+}
+trap remove_docker_container EXIT
+
+# Generate corresponding --device args based on BUILDKITE_AGENT_NAME
+# Ascend NPU BUILDKITE_AGENT_NAME format is {hostname}-{agent_idx}-{npu_card_num}cards, and agent_idx starts from 1.
+#   e.g. atlas-a2-001-1-2cards means this is the 1-th agent on atlas-a2-001 host, and it has 2 NPU cards.
+#   returns --device /dev/davinci0 --device /dev/davinci1
+parse_and_gen_devices() {
+    local input="$1"
+    local index cards_num
+    if [[ "$input" =~ ([0-9]+)-([0-9]+)cards$ ]]; then
+        index="${BASH_REMATCH[1]}"
+        cards_num="${BASH_REMATCH[2]}"
+    else
+        echo "parse error" >&2
+        return 1
+    fi
+
+    local devices=""
+    local i=0
+    while (( i < cards_num )); do
+        local dev_idx=$(((index - 1)*cards_num + i ))
+        devices="$devices --device /dev/davinci${dev_idx}"
+        ((i++))
+    done
+
+    # trim leading space
+    devices="${devices#"${devices%%[![:space:]]*}"}"
+    # Output devices: assigned to the caller variable
+    printf '%s' "$devices"
+}
+
+devices=$(parse_and_gen_devices "${BUILDKITE_AGENT_NAME}") || exit 1
+
+# Run the image and execute the Out-Of-Tree (OOT) platform interface test case on Ascend NPU hardware.
+# This test checks whether the OOT platform interface is functioning properly in conjunction with
+# the hardware plugin vllm-ascend.
+hf_model_cache_dir=/mnt/hf_cache${agent_idx}
+ms_model_cache_dir=/mnt/modelscope${agent_idx}
+mkdir -p ${hf_model_cache_dir}
+mkdir -p ${ms_model_cache_dir}
+docker run \
+    --init \
+    ${devices} \
+    --device /dev/davinci_manager \
+    --device /dev/devmm_svm \
+    --device /dev/hisi_hdc \
+    -v /usr/local/dcmi:/usr/local/dcmi \
+    -v /usr/local/bin/npu-smi:/usr/local/bin/npu-smi \
+    -v /usr/local/Ascend/driver/lib64/:/usr/local/Ascend/driver/lib64/ \
+    -v /usr/local/Ascend/driver/version.info:/usr/local/Ascend/driver/version.info \
+    -v /etc/ascend_install.info:/etc/ascend_install.info \
+    -v ${hf_model_cache_dir}:/root/.cache/huggingface \
+    -v ${ms_model_cache_dir}:/root/.cache/modelscope \
+    --network host \
+    --entrypoint="" \
+    --name "${container_name}" \
+    "${image_name}" \
+    bash -c '
+    set -e
+    VLLM_USE_MODELSCOPE=True pytest -s -v tests/e2e/offline_inference/test_qwen2_5_omni_expansion.py
+'

.buildkite/scripts/upload_pipeline.py

@@ -0,0 +1,499 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: Apache-2.0
+"""Render and optionally upload Buildkite pipeline YAML with diff-aware logic.
+
+Bootstrap mode (pipeline.yml with __IMAGE_BUILD_IF__ placeholders):
+  - Detect docs-only, pytest skip-mark-only, or combined docs+skip-mark PR/main changes and
+    substitute skip-ci ``if`` expressions.
+
+Test pipeline mode (e.g. test-merge.yml):
+  - Drop steps or groups whose ``source_file_dependencies`` do not match changed files.
+  - ``source_file_dependencies`` may be set on a leaf step or on a group step.
+  - Always strip ``source_file_dependencies`` from uploaded YAML (Buildkite does not
+    recognize this uploader-only key).
+
+Usage:
+  python3 upload_pipeline.py [--upload] <pipeline.yml>
+
+  # Bootstrap (replaces upload_pipeline_with_skip_ci.sh):
+  python3 upload_pipeline.py --upload .buildkite/pipeline.yml
+
+  # Test pipeline (replaces upload_test_pipeline_with_diff_skip.py):
+  python3 upload_pipeline.py --upload .buildkite/test-merge.yml
+"""
+
+from __future__ import annotations
+
+import argparse
+import os
+import re
+import subprocess
+import sys
+from pathlib import Path
+from typing import Any
+
+try:
+    import yaml
+except ImportError:
+    subprocess.check_call(
+        [sys.executable, "-m", "pip", "install", "-q", "pyyaml"],
+    )
+    import yaml
+
+LOG = "upload_pipeline"
+ROOT = Path(__file__).resolve().parent.parent.parent
+DOC_SEP = "\n---\n"
+BOOTSTRAP_MARKER = "__IMAGE_BUILD_IF__"
+
+
+def _log(message: str) -> None:
+    print(f"{LOG}: {message}", file=sys.stderr)
+
+
+def _git(*args: str) -> subprocess.CompletedProcess[str]:
+    return subprocess.run(
+        ["git", *args],
+        check=False,
+        capture_output=True,
+        text=True,
+    )
+
+
+def resolve_diff_range() -> str | None:
+    """Return a git diff range for PR or main builds, or None when unavailable."""
+    is_pr = os.environ.get("BUILDKITE_PULL_REQUEST", "false") != "false" and os.environ.get(
+        "BUILDKITE_PULL_REQUEST", ""
+    )
+    commit = os.environ.get("BUILDKITE_COMMIT", "")
+
+    if is_pr:
+        base_branch = os.environ.get("BUILDKITE_PULL_REQUEST_BASE_BRANCH", "main")
+        base_ref = f"origin/{base_branch}"
+        if _git("rev-parse", "--verify", base_ref).returncode != 0:
+            _log(f"origin/{base_branch} not found locally; trying fetch")
+            _git("fetch", "--depth=200", "origin", base_branch)
+        if _git("rev-parse", "--verify", base_ref).returncode != 0:
+            if _git("rev-parse", "--verify", base_branch).returncode == 0:
+                base_ref = base_branch
+            else:
+                _log(f"cannot resolve PR base {base_branch}; using safe defaults")
+                return None
+        return f"{base_ref}...{commit}"
+
+    if os.environ.get("BUILDKITE_BRANCH", "") == "main":
+        if _git("rev-parse", "--verify", f"{commit}^").returncode != 0:
+            _log("main commit has no parent; using safe defaults")
+            return None
+        return f"{commit}^..{commit}"
+
+    _log("not PR/main build; using safe defaults")
+    return None
+
+
+def resolve_changed_files() -> list[str] | None:
+    """Return changed file paths, or None when diff cannot be resolved."""
+    diff_range = resolve_diff_range()
+    if diff_range is None:
+        return None
+
+    result = _git("diff", "--name-only", diff_range)
+    if result.returncode != 0:
+        _log(f"git diff failed ({diff_range}); using safe defaults")
+        return None
+
+    files = [line for line in result.stdout.splitlines() if line.strip()]
+    _log(f"{len(files)} changed file(s)")
+    return files
+
+
+def _is_doc_file(file_path: str) -> bool:
+    if not file_path:
+        return False
+    if file_path.startswith("docs/"):
+        return True
+    if file_path.endswith(".md"):
+        return True
+    return file_path == "mkdocs.yml"
+
+
+def is_docs_only_change(changed_files: list[str]) -> bool:
+    has_any = False
+    for file_path in changed_files:
+        if not file_path:
+            continue
+        has_any = True
+        if not _is_doc_file(file_path):
+            return False
+    return has_any
+
+
+def _is_test_python_file(file_path: str) -> bool:
+    path = Path(file_path)
+    return path.suffix == ".py" and bool(path.parts) and path.parts[0] == "tests"
+
+
+_SKIP_MARK_RE = re.compile(r"pytest\.mark\.skip(?:if)?\b|pytest\.skip\s*\(")
+_PYTESTMARK_SKIP_RE = re.compile(r"pytest\.mark\.skip\b")
+_PYTEST_MARK_ONLY_RE = re.compile(r"pytest\.mark\.\w+")
+
+
+def _paren_balance(line: str) -> int:
+    return line.count("(") - line.count(")")
+
+
+def _is_skip_mark_related_content(line: str) -> bool:
+    stripped = line.strip()
+    if not stripped:
+        return True
+    if stripped.startswith("#"):
+        return True
+    return _SKIP_MARK_RE.search(stripped) is not None
+
+
+def _is_pytestmark_adjacent_content(line: str) -> bool:
+    """Allow pytestmark list refactors that only add skip/skipif alongside existing marks."""
+    stripped = line.strip().rstrip(",")
+    if not stripped:
+        return True
+    if stripped in {"[", "]"}:
+        return True
+    if stripped.startswith("#"):
+        return True
+    if stripped.startswith("pytestmark"):
+        return True
+    return _PYTEST_MARK_ONLY_RE.search(stripped) is not None
+
+
+def diff_only_contains_skip_mark_changes(diff_text: str) -> bool:
+    """True when diff only edits skip marks or reformats pytestmark to add them."""
+    pending_depth = 0
+    saw_change = False
+    has_skip_mark_edit = False
+    for raw_line in diff_text.splitlines():
+        if raw_line.startswith("@@"):
+            pending_depth = 0
+            continue
+        if not (raw_line.startswith("+") or raw_line.startswith("-")):
+            continue
+        if raw_line.startswith("+++") or raw_line.startswith("---"):
+            continue
+
+        saw_change = True
+        content = raw_line[1:]
+
+        if pending_depth > 0:
+            pending_depth += _paren_balance(content)
+            if pending_depth < 0:
+                pending_depth = 0
+            continue
+
+        if _is_skip_mark_related_content(content):
+            has_skip_mark_edit = True
+            pending_depth = max(0, _paren_balance(content))
+            continue
+
+        if _is_pytestmark_adjacent_content(content):
+            continue
+
+        return False
+
+    return saw_change and has_skip_mark_edit
+
+
+def _git_file_diff(file_path: str, diff_range: str) -> str | None:
+    result = _git("diff", diff_range, "--", file_path)
+    if result.returncode != 0:
+        _log(f"skip-mark-only: git diff failed for {file_path}")
+        return None
+    return result.stdout
+
+
+def _is_new_file_diff(diff_text: str) -> bool:
+    return "new file mode" in diff_text or "--- /dev/null" in diff_text
+
+
+def _file_has_module_level_pytest_skip(content: str) -> bool:
+    """True when ``pytestmark`` applies unconditional ``pytest.mark.skip`` to the module."""
+    match = re.search(r"^pytestmark\s*=\s*\[(.*?)^\s*\]", content, re.MULTILINE | re.DOTALL)
+    if match is not None:
+        return _PYTESTMARK_SKIP_RE.search(match.group(1)) is not None
+    match = re.search(r"^pytestmark\s*=\s*pytest\.mark\.skip\b", content, re.MULTILINE)
+    return match is not None
+
+
+def _file_content_at_commit(file_path: str, commit: str) -> str | None:
+    result = _git("show", f"{commit}:{file_path}")
+    if result.returncode != 0:
+        _log(f"skip-mark-only: cannot read {file_path} at {commit}")
+        return None
+    return result.stdout
+
+
+def _qualifies_as_skip_mark_test_change(file_path: str, diff_range: str) -> bool:
+    diff_text = _git_file_diff(file_path, diff_range)
+    if diff_text is None:
+        return False
+    if not diff_text.strip():
+        _log(f"skip-mark-only: empty diff for {file_path}")
+        return False
+
+    if diff_only_contains_skip_mark_changes(diff_text):
+        return True
+
+    if not _is_new_file_diff(diff_text):
+        _log(f"skip-mark-only: non-skip changes in {file_path}")
+        return False
+
+    commit = os.environ.get("BUILDKITE_COMMIT", "")
+    if not commit:
+        return False
+    content = _file_content_at_commit(file_path, commit)
+    if content is None:
+        return False
+    if _file_has_module_level_pytest_skip(content):
+        _log(f"skip-mark-only: new test file with module-level pytest.mark.skip: {file_path}")
+        return True
+
+    _log(f"skip-mark-only: new test file without module-level pytest.mark.skip: {file_path}")
+    return False
+
+
+def is_skip_mark_only_change(changed_files: list[str], *, diff_range: str) -> bool:
+    """True when every changed file is a tests/ module with only skip-mark edits."""
+    has_any = False
+    for file_path in changed_files:
+        if not file_path:
+            continue
+        has_any = True
+        if not _is_test_python_file(file_path):
+            return False
+        if not _qualifies_as_skip_mark_test_change(file_path, diff_range):
+            return False
+    return has_any
+
+
+def is_docs_or_skip_mark_only_change(changed_files: list[str], *, diff_range: str) -> bool:
+    """True when each changed file is a doc path or a qualifying skip-mark test change."""
+    has_any = False
+    for file_path in changed_files:
+        if not file_path:
+            continue
+        has_any = True
+        if _is_doc_file(file_path):
+            continue
+        if _is_test_python_file(file_path) and _qualifies_as_skip_mark_test_change(file_path, diff_range):
+            continue
+        _log(f"docs/skip-mark-only: rejecting {file_path}")
+        return False
+    return has_any
+
+
+def resolve_skip_ci(changed_files: list[str] | None, *, diff_range: str | None = None) -> bool:
+    if changed_files is None:
+        _log("skip-ci=0 (could not resolve changed files)")
+        return False
+
+    if diff_range is not None and is_docs_or_skip_mark_only_change(changed_files, diff_range=diff_range):
+        if is_docs_only_change(changed_files):
+            _log("docs-only change detected; skip-ci=1")
+        elif is_skip_mark_only_change(changed_files, diff_range=diff_range):
+            _log("pytest skip-mark-only change detected; skip-ci=1")
+        else:
+            _log("docs + pytest skip-mark-only change detected; skip-ci=1")
+        return True
+
+    if diff_range is None and is_docs_only_change(changed_files):
+        _log("docs-only change detected; skip-ci=1")
+        return True
+
+    _log("non-doc/non-skip-mark changes detected; skip-ci=0")
+    return False
+
+
+def render_bootstrap_pipeline(text: str, *, skip_ci: bool) -> str:
+    if DOC_SEP in text:
+        _, continuation = text.split(DOC_SEP, 1)
+    else:
+        continuation = text
+
+    nightly_only = (
+        '(build.pull_request.labels includes "nightly-test") || (build.branch == "main" && build.env("NIGHTLY") == "1")'
+    )
+    if skip_ci:
+        image_if = f"'{nightly_only}'"
+        ready_if = "'false'"
+        merge_if = "'false'"
+    else:
+        image_if = "'true'"
+        ready_if = '\'build.branch != "main" && build.pull_request.labels includes "ready"\''
+        merge_if = (
+            '\'(build.branch == "main" && build.env("NIGHTLY") != "1" '
+            '&& build.env("WEEKLY") != "1") || '
+            '(build.branch != "main" && build.pull_request.labels includes "merge-test")\''
+        )
+
+    return (
+        continuation.replace("__IMAGE_BUILD_IF__", image_if)
+        .replace("__UPLOAD_READY_IF__", ready_if)
+        .replace("__UPLOAD_MERGE_IF__", merge_if)
+    )
+
+
+def _matches_dependencies(changed_files: list[str], prefixes: list[str]) -> bool:
+    for path in changed_files:
+        for prefix in prefixes:
+            normalized = prefix.rstrip("/")
+            if path == normalized or path.startswith(f"{normalized}/"):
+                return True
+    return False
+
+
+def _strip_source_file_dependencies(step: dict[str, Any]) -> dict[str, Any]:
+    return {key: value for key, value in step.items() if key != "source_file_dependencies"}
+
+
+def _step_label(step: dict[str, Any]) -> str:
+    return str(step.get("group") or step.get("label") or "<step>")
+
+
+def _filter_steps(steps: list[Any], changed_files: list[str]) -> list[Any]:
+    filtered: list[Any] = []
+    for step in steps:
+        if not isinstance(step, dict):
+            filtered.append(step)
+            continue
+
+        deps = step.get("source_file_dependencies")
+        if deps is not None and not isinstance(deps, list):
+            raise ValueError(
+                f"source_file_dependencies must be a list in step {_step_label(step)!r}",
+            )
+        if deps is not None and not _matches_dependencies(changed_files, deps):
+            _log(
+                f"skip {_step_label(step)!r} (no changes under {deps})",
+            )
+            continue
+
+        nested = step.get("steps")
+        if nested is not None:
+            kept_nested = _filter_steps(nested, changed_files)
+            if not kept_nested:
+                _log(f"omit empty group {_step_label(step)!r}")
+                continue
+            new_step = _strip_source_file_dependencies(step)
+            new_step["steps"] = kept_nested
+            filtered.append(new_step)
+            continue
+
+        if deps is not None:
+            filtered.append(_strip_source_file_dependencies(step))
+        else:
+            filtered.append(step)
+
+    return filtered
+
+
+def _strip_uploader_metadata_from_steps(steps: list[Any]) -> list[Any]:
+    """Remove uploader-only keys while keeping all steps (no diff filtering)."""
+    stripped: list[Any] = []
+    for step in steps:
+        if not isinstance(step, dict):
+            stripped.append(step)
+            continue
+
+        deps = step.get("source_file_dependencies")
+        if deps is not None and not isinstance(deps, list):
+            raise ValueError(
+                f"source_file_dependencies must be a list in step {_step_label(step)!r}",
+            )
+
+        nested = step.get("steps")
+        new_step = _strip_source_file_dependencies(step)
+        if nested is not None:
+            new_step["steps"] = _strip_uploader_metadata_from_steps(nested)
+        stripped.append(new_step)
+
+    return stripped
+
+
+def render_test_pipeline(
+    doc: dict[str, Any],
+    changed_files: list[str] | None,
+) -> dict[str, Any]:
+    steps = doc.get("steps")
+    if not isinstance(steps, list):
+        return doc
+    if changed_files is not None:
+        steps = _filter_steps(steps, changed_files)
+    else:
+        steps = _strip_uploader_metadata_from_steps(steps)
+    return {**doc, "steps": steps}
+
+
+def resolve_pipeline_path(arg: str) -> Path:
+    path = Path(arg)
+    if path.is_absolute():
+        return path
+    return ROOT / path
+
+
+def render_pipeline(path: Path) -> str:
+    text = path.read_text(encoding="utf-8")
+    diff_range = resolve_diff_range()
+    changed_files = resolve_changed_files()
+
+    if BOOTSTRAP_MARKER in text:
+        rendered = render_bootstrap_pipeline(
+            text,
+            skip_ci=resolve_skip_ci(changed_files, diff_range=diff_range),
+        )
+        return rendered
+
+    doc = yaml.safe_load(text)
+    if not isinstance(doc, dict):
+        raise ValueError(f"invalid pipeline YAML: {path}")
+
+    doc = render_test_pipeline(doc, changed_files)
+
+    return yaml.safe_dump(doc, sort_keys=False)
+
+
+def upload_to_buildkite(content: str) -> None:
+    subprocess.run(
+        ["buildkite-agent", "pipeline", "upload"],
+        input=content,
+        text=True,
+        check=True,
+    )
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument(
+        "pipeline",
+        nargs="?",
+        default=".buildkite/pipeline.yml",
+        help="Pipeline YAML path (default: .buildkite/pipeline.yml)",
+    )
+    parser.add_argument(
+        "--upload",
+        action="store_true",
+        help="Pipe rendered YAML to buildkite-agent pipeline upload",
+    )
+    args = parser.parse_args()
+
+    path = resolve_pipeline_path(args.pipeline)
+    if not path.is_file():
+        _log(f"missing pipeline file: {path}")
+        return 1
+
+    rendered = render_pipeline(path)
+    if args.upload:
+        upload_to_buildkite(rendered)
+    else:
+        sys.stdout.write(rendered)
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

.buildkite/test-amd-merge.yml

@@ -7,29 +7,17 @@ steps:
 
 - group: ":card_index_dividers: Simple Test"
   steps:
-    - label: "Simple · Diffusion & Model Executor Test (1/2)"
+    - label: "Simple · Diffusion & Model Executor Test"
       agent_pool: mi325_1
       depends_on: amd-build
-      timeout_in_minutes: 70
+      timeout_in_minutes: 50
       mirror_hardwares: [amdproduction]
       grade: Blocking
       commands:
         - export VLLM_ROCM_USE_AITER=0
         # ignore test_teacache_extractors.py because it use rocm gemm kernel from vLLM
         # that is not supported on CPU
-        # CPU tests on ROCm gpu machines are very slow, so we give them more time.
-        - "timeout 60m pytest -sv tests/diffusion -m 'core_model and cpu' --ignore=tests/diffusion/cache/test_teacache_extractors.py --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
-
-    - label: "Simple · Diffusion & Model Executor Test (2/2)"
-      agent_pool: mi325_1
-      depends_on: amd-build
-      timeout_in_minutes: 70
-      mirror_hardwares: [amdproduction]
-      grade: Blocking
-      commands:
-        - export VLLM_ROCM_USE_AITER=0
-        # CPU tests on ROCm gpu machines are very slow, so we give them more time.
-        - "timeout 60m pytest -sv tests/model_executor -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
+        - "timeout 40m pytest -sv tests/diffusion tests/model_executor -m 'core_model and cpu' --ignore=tests/diffusion/cache/test_teacache_extractors.py --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
 
     - label: "Simple · Other Test"
       agent_pool: mi325_1
@@ -89,7 +77,7 @@ steps:
       mirror_hardwares: [amdproduction]
       grade: Blocking
       commands:
-        - pytest -s -v tests/diffusion/distributed/test_tensor_parallel.py
+        - pytest -s -v tests/diffusion/distributed/test_zimage_parallelism.py
 
     - label: "Diffusion Model Test"
       agent_pool: mi325_1
@@ -170,7 +158,7 @@ steps:
       grade: Blocking
       commands:
         - |
-          timeout 50m bash -c '
+          timeout 20m bash -c '
             export VLLM_LOGGING_LEVEL=DEBUG
             export VLLM_WORKER_MULTIPROC_METHOD=spawn
             export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
@@ -182,14 +170,39 @@ steps:
       depends_on: amd-build
       mirror_hardwares: [amdproduction]
       grade: Blocking
-      timeout_in_minutes: 50
       commands:
         - |
-          timeout 40m bash -c '
+          timeout 30m bash -c '
             export VLLM_LOGGING_LEVEL=DEBUG
             export VLLM_WORKER_MULTIPROC_METHOD=spawn
             export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
-            pytest -s -v tests/e2e/online_serving/test_qwen3_tts_base.py tests/e2e/offline_inference/test_qwen3_tts_base.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            pytest -s -v tests/e2e/online_serving/test_qwen3_tts_base.py tests/e2e/offline_inference/test_qwen3_tts_base.py -m "advanced_model" --run-level "advanced_model"
+          '
+
+    # It is the whisper ASR validation step that is slow, not the Voxtral TTS generation.
+    # whisper ASR runs on CPU.
+    - label: "Voxtral-TTS E2E Test (Online Serving)"
+      agent_pool: mi325_1
+      depends_on: amd-build
+      mirror_hardwares: [amdproduction]
+      grade: Blocking
+      commands:
+        - |
+          timeout 40m bash -c '
+            export VLLM_WORKER_MULTIPROC_METHOD=spawn
+            pytest -s -v tests/e2e/online_serving/test_voxtral_tts.py -m "advanced_model" --run-level "advanced_model"
+          '
+
+    - label: "Voxtral-TTS E2E Test (Offline Inference)"
+      agent_pool: mi325_1
+      depends_on: amd-build
+      mirror_hardwares: [amdproduction]
+      grade: Blocking
+      commands:
+        - |
+          timeout 30m bash -c '
+            export VLLM_WORKER_MULTIPROC_METHOD=spawn
+            pytest -s -v tests/e2e/offline_inference/test_voxtral_tts.py -m "advanced_model" --run-level "advanced_model"
           '
 
 # TODO: Bagel test on ROCm is very unstable. @tjtanaa

.buildkite/test-amd-ready.yaml

@@ -35,7 +35,7 @@ steps:
   mirror_hardwares: [amdproduction]
   grade: Blocking
   commands:
-    - timeout 20m pytest -s -v tests/e2e/features/custom_pipeline/ -m "core_model"
+    - timeout 20m pytest -s -v tests/e2e/offline_inference/custom_pipeline/ -m "core_model"
 
 
 - label: "Voxtral TTS CUDA Unit Test"
@@ -161,6 +161,19 @@ steps:
 
 - group: ":card_index_dividers: E2E Test"
   steps:
+    - label: "MiMo-Audio E2E Test with H100"
+      agent_pool: mi325_1
+      depends_on: amd-build
+      mirror_hardwares: [amdproduction]
+      grade: Blocking
+      commands:
+        - |
+          timeout 30m bash -c '
+            export VLLM_LOGGING_LEVEL=DEBUG
+            export VLLM_WORKER_MULTIPROC_METHOD=spawn
+            pytest -s -v tests/e2e/online_serving/test_mimo_audio.py -m "core_model" --run-level "core_model"
+          '
+
     - label: "Qwen3-TTS E2E Test"
       agent_pool: mi325_1
       depends_on: amd-build
@@ -180,7 +193,7 @@ steps:
       commands:
         - |
           timeout 20m bash -c '
-            pytest -s -v tests/e2e/online_serving/test_cosyvoice3_tts_expansion.py -m "slow" --run-level "core_model"
+            pytest -s -v tests/e2e/online_serving/test_cosyvoice3_tts_expansion.py -m "full_model" --run-level "core_model"
           '
 
 # TODO: Bagel test on ROCm is very unstable. @tjtanaa

.buildkite/test-amd.yaml

@@ -0,0 +1,118 @@
+steps:
+
+- label: "Diffusion Model Test"
+  timeout_in_minutes: 30
+  agent_pool: mi325_2
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - export GPU_ARCHS=gfx942
+    - pytest -s -v tests/e2e/offline_inference/test_z_image.py
+
+- label: "Diffusion Images API LoRA E2E"
+  timeout_in_minutes: 30
+  agent_pool: mi325_1
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - export GPU_ARCHS=gfx942
+    - export VLLM_LOGGING_LEVEL=DEBUG
+    - export VLLM_WORKER_MULTIPROC_METHOD=spawn
+    - pytest -s -v tests/diffusion/lora/test_images_generations_lora.py
+
+- label: "Diffusion Model CPU offloading Test"
+  timeout_in_minutes: 20
+  agent_pool: mi325_1
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - export GPU_ARCHS=gfx942
+    - export VLLM_LOGGING_LEVEL=DEBUG
+    - export VLLM_WORKER_MULTIPROC_METHOD=spawn
+    - pytest -s -v tests/diffusion/offloader/test_diffusion_cpu_offload.py
+    - pytest -s -v tests/diffusion/offloader/test_diffusion_layerwise_offload.py
+
+- label: "Diffusion Cache Backend Test"
+  timeout_in_minutes: 15
+  agent_pool: mi325_1
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - export GPU_ARCHS=gfx942
+    - export VLLM_LOGGING_LEVEL=DEBUG
+    - export VLLM_WORKER_MULTIPROC_METHOD=spawn
+    - pytest -s -v -m 'core_model and cache and diffusion and not distributed_rocm and MI325'
+
+- label: "Diffusion Sequence Parallelism Test"
+  timeout_in_minutes: 20
+  agent_pool: mi325_2
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - export GPU_ARCHS=gfx942
+    - export VLLM_LOGGING_LEVEL=DEBUG
+    - export VLLM_WORKER_MULTIPROC_METHOD=spawn
+    - pytest -s -v tests/diffusion/distributed/test_sequence_parallel.py -m core_model
+
+- label: "Diffusion Tensor Parallelism Test"
+  timeout_in_minutes: 20
+  agent_pool: mi325_2
+  depends_on: amd-build
+  commands:
+    - export GPU_ARCHS=gfx942
+    - export VLLM_LOGGING_LEVEL=DEBUG
+    - export VLLM_WORKER_MULTIPROC_METHOD=spawn
+    - pytest -s -v tests/diffusion/distributed/test_zimage_parallelism.py
+
+- label: "Diffusion GPU Worker Test"
+  timeout_in_minutes: 20
+  agent_pool: mi325_2
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - pytest -s -v tests/diffusion/test_diffusion_worker.py
+
+- label: "Omni Model Test Qwen2-5-Omni"
+  timeout_in_minutes: 15
+  agent_pool: mi325_2
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - export GPU_ARCHS=gfx942
+    - export VLLM_LOGGING_LEVEL=DEBUG
+    - export VLLM_WORKER_MULTIPROC_METHOD=spawn
+    - pytest -s -v tests/e2e/offline_inference/test_qwen2_5_omni_expansion.py
+    - pytest -s -v tests/engine/test_async_omni_engine_abort.py
+
+- label: "Omni Model Test Qwen3-Omni"
+  timeout_in_minutes: 15
+  agent_pool: mi325_2
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - export VLLM_LOGGING_LEVEL=DEBUG
+    - export VLLM_WORKER_MULTIPROC_METHOD=spawn
+    - export VLLM_ROCM_USE_AITER=0
+    - pytest -s -v tests/e2e/offline_inference/test_qwen3_omni.py
+    - pytest -s -v tests/e2e/online_serving/test_qwen3_omni.py
+
+
+- label: "Omni Sleep Mode Test"
+  timeout_in_minutes: 40
+  agent_pool: mi325_2
+  depends_on: amd-build
+  mirror_hardwares: [amdproduction]
+  grade: Blocking
+  commands:
+    - export GPU_ARCHS=gfx942
+    - export VLLM_LOGGING_LEVEL=DEBUG
+    - export VLLM_WORKER_MULTIPROC_METHOD=spawn
+    - pytest -s -v tests/entrypoints/test_omni_sleep_mode.py -m "advanced_model and omni and MI325" --run-level "advanced_model"

.buildkite/test-merge.yml

@@ -0,0 +1,943 @@
+env:
+  VLLM_WORKER_MULTIPROC_METHOD: spawn
+  VLLM_USE_FLASHINFER_MOE_FP16: 0
+  HF_HUB_DOWNLOAD_TIMEOUT: 300
+  HF_HUB_ETAG_TIMEOUT: 60
+
+steps:
+  - group: ":card_index_dividers: Simple Test"
+    depends_on: upload-merge-pipeline
+    steps:
+      - label: "Simple · Diffusion & Model Executor Test"
+        commands:
+          - "timeout 20m pytest -sv tests/diffusion tests/model_executor -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Simple · Other Test"
+        commands:
+          - "timeout 20m pytest -sv -m 'core_model and cpu' --ignore=tests/diffusion --ignore=tests/model_executor --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+  - group: ":card_index_dividers: Diffusion Test"
+    depends_on: upload-merge-pipeline
+    steps:
+      - label: "Diffusion · LoRA Test"
+        timeout_in_minutes: 30
+        commands:
+          - pytest -sv tests/diffusion/lora/ -m 'advanced_model and cuda' --run-level "advanced_model"
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+  - label: "Distributed Test"
+    timeout_in_minutes: 60
+    commands:
+      - |
+        timeout 55m bash -c "
+          set -e
+          export VLLM_IMAGE_FETCH_TIMEOUT=60
+          pytest -s -v tests/distributed -m 'advanced_model and cuda' --run-level 'advanced_model'
+        "
+    agents:
+      queue: "mithril-h100-pool"
+    plugins:
+      - kubernetes:
+          podSpec:
+            containers:
+              - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                resources:
+                  limits:
+                    nvidia.com/gpu: 1
+                volumeMounts:
+                  - name: devshm
+                    mountPath: /dev/shm
+                  - name: hf-cache
+                    mountPath: /root/.cache/huggingface
+                env:
+                  - name: HF_HOME
+                    value: /root/.cache/huggingface
+                  - name: HF_TOKEN
+                    valueFrom:
+                      secretKeyRef:
+                        name: hf-token-secret
+                        key: token
+            nodeSelector:
+              node.kubernetes.io/instance-type: gpu-h100-sxm
+            volumes:
+              - name: devshm
+                emptyDir:
+                  medium: Memory
+              - name: hf-cache
+                hostPath:
+                  path: /mnt/hf-cache
+                  type: DirectoryOrCreate
+
+  - group: ":card_index_dividers: Entrypoints Test"
+    depends_on: upload-merge-pipeline
+    steps:
+      - label: "Entrypoint Test with H100"
+        timeout_in_minutes: 60
+        commands:
+          - pytest -s -v tests/entrypoints/ -m "advanced_model and cuda and H100" --run-level "advanced_model"
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 2
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Entrypoint Test with L4"
+        timeout_in_minutes: 30
+        commands:
+          - pytest -s -v tests/entrypoints/ -m "advanced_model and cuda and L4" --run-level "advanced_model"
+        agents:
+          queue: "gpu_4_queue" # g6.12xlarge instance on AWS, has 4 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+  - group: ":card_index_dividers: E2E Test"
+    depends_on: upload-merge-pipeline
+    steps:
+      - label: "TTS · Step-Audio2 E2E Test"
+        timeout_in_minutes: 20
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              pytest -s -v tests/e2e/offline_inference/test_step_audio2.py -m 'advanced_model' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "gpu_4_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "TTS · Qwen3-TTS CustomVoice Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen3_tts_customvoice.py
+          - tests/e2e/offline_inference/test_qwen3_tts_customvoice.py
+          - vllm_omni/model_executor/models/qwen3_tts/
+          - vllm_omni/model_executor/stage_input_processors/qwen3_tts.py
+          - vllm_omni/deploy/qwen3_tts.yaml
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
+              pytest -s -v tests/e2e/online_serving/test_qwen3_tts_customvoice.py tests/e2e/offline_inference/test_qwen3_tts_customvoice.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "TTS · Qwen3-TTS Base Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen3_tts_base.py
+          - tests/e2e/offline_inference/test_qwen3_tts_base.py
+          - vllm_omni/model_executor/models/qwen3_tts/
+          - vllm_omni/model_executor/stage_input_processors/qwen3_tts.py
+          - vllm_omni/deploy/qwen3_tts.yaml
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
+              pytest -s -v tests/e2e/online_serving/test_qwen3_tts_base.py tests/e2e/offline_inference/test_qwen3_tts_base.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "TTS · Ming-Omni-TTS Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_ming_tts.py
+          - tests/e2e/offline_inference/test_ming_tts.py
+          - vllm_omni/model_executor/models/ming_tts/
+          - vllm_omni/model_executor/stage_input_processors/ming_tts.py
+          - vllm_omni/deploy/ming_tts.yaml
+        timeout_in_minutes: 20
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              pytest -s -v tests/e2e/online_serving/test_ming_tts.py tests/e2e/offline_inference/test_ming_tts.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "TTS · Voxtral-TTS Test"
+        source_file_dependencies:
+          - tests/e2e/offline_inference/test_voxtral_tts.py
+          - tests/e2e/online_serving/test_voxtral_tts.py
+          - vllm_omni/model_executor/models/voxtral_tts/
+          - vllm_omni/model_executor/stage_input_processors/voxtral_tts.py
+          - vllm_omni/deploy/voxtral_tts.yaml
+        timeout_in_minutes: 20
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              pytest -s -v tests/e2e/offline_inference/test_voxtral_tts.py tests/e2e/online_serving/test_voxtral_tts.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "TTS · GLM-TTS Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_glm_tts.py
+          - tests/e2e/offline_inference/test_glm_tts.py
+          - vllm_omni/model_executor/models/glm_tts/
+          - vllm_omni/model_executor/stage_input_processors/glm_tts.py
+          - vllm_omni/deploy/glm_tts.yaml
+        timeout_in_minutes: 20
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              pytest -s -v tests/e2e/online_serving/test_glm_tts.py tests/e2e/offline_inference/test_glm_tts.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Omni · Qwen3-Omni Test"
+        source_file_dependencies:
+          - tests/e2e/offline_inference/test_qwen3_omni.py
+          - tests/e2e/online_serving/test_qwen3_omni.py
+          - vllm_omni/model_executor/models/qwen3_omni/
+          - vllm_omni/model_executor/stage_input_processors/qwen3_omni.py
+          - vllm_omni/deploy/qwen3_omni_moe.yaml
+        timeout_in_minutes: 30
+        commands:
+          - pytest -s -v tests/e2e/offline_inference/test_qwen3_omni.py tests/e2e/online_serving/test_qwen3_omni.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 2
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Omni · Mimo Audio Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_mimo_audio.py
+          - vllm_omni/model_executor/models/mimo_audio/
+          - vllm_omni/model_executor/stage_input_processors/mimo_audio.py
+          - vllm_omni/deploy/mimo_audio.yaml
+        timeout_in_minutes: 30
+        commands:
+          - pytest -s -v tests/e2e/online_serving/test_mimo_audio.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 2
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "TTS · VoxCPM2 Test"
+        source_file_dependencies:
+          - tests/e2e/offline_inference/test_voxcpm2_tts.py
+          - tests/e2e/online_serving/test_voxcpm2_tts.py
+          - vllm_omni/model_executor/models/voxcpm2/
+          - vllm_omni/deploy/voxcpm2.yaml
+        timeout_in_minutes: 30
+        commands:
+          - |
+            timeout 30m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              pytest -s -v tests/e2e/offline_inference/test_voxcpm2_tts.py tests/e2e/online_serving/test_voxcpm2_tts.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Omni · Higgs-Audio V2 Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_higgs_audio_v2.py
+          - tests/e2e/offline_inference/test_higgs_audio_v2.py
+          - vllm_omni/model_executor/models/higgs_audio_v2/
+          - vllm_omni/model_executor/stage_input_processors/higgs_audio_v2.py
+          - vllm_omni/deploy/higgs_audio_v2.yaml
+        timeout_in_minutes: 30
+        commands:
+          - |
+            timeout 30m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              export VLLM_USE_DEEP_GEMM=0
+              export VLLM_MOE_USE_DEEP_GEMM=0
+              pytest -s -v tests/e2e/online_serving/test_higgs_audio_v2.py tests/e2e/offline_inference/test_higgs_audio_v2.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · Qwen Image Edit Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen_image_edit.py
+          - vllm_omni/diffusion/models/qwen_image/
+        timeout_in_minutes: 20
+        commands:
+          - pytest -s -v tests/e2e/online_serving/test_qwen_image_edit.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Longcat Image Edit Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_longcat_image_edit.py
+          - vllm_omni/diffusion/models/longcat_image/
+        timeout_in_minutes: 20
+        commands:
+          - pytest -s -v tests/e2e/online_serving/test_longcat_image_edit.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Longcat Image Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_longcat_image.py
+          - vllm_omni/diffusion/models/longcat_image/
+        timeout_in_minutes: 20
+        commands:
+          - pytest -s -v tests/e2e/online_serving/test_longcat_image.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Bagel Test"
+        source_file_dependencies:
+          - tests/e2e/offline_inference/test_bagel.py
+          - tests/e2e/online_serving/test_bagel.py
+          - vllm_omni/model_executor/models/bagel/
+          - vllm_omni/diffusion/models/bagel/
+          - vllm_omni/model_executor/stage_input_processors/bagel.py
+          - vllm_omni/deploy/bagel.yaml
+        timeout_in_minutes: 20
+        commands:
+          - pytest -s -v tests/e2e/offline_inference/test_bagel.py tests/e2e/online_serving/test_bagel.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+
+      - label: "Diffusion · Qwen Image Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen_image.py
+          - vllm_omni/diffusion/models/qwen_image/
+        commands:
+          - |
+            timeout 40m bash -c '
+              export VLLM_IMAGE_FETCH_TIMEOUT=60
+              pytest -s -v tests/e2e/online_serving/test_qwen_image.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            '
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Qwen Image Layered Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen_image_layered.py
+          - vllm_omni/diffusion/models/qwen_image/
+        commands:
+          - |
+            timeout 40m bash -c "
+              export VLLM_IMAGE_FETCH_TIMEOUT=60
+              pytest -s -v tests/e2e/online_serving/test_qwen_image_layered.py -m 'advanced_model and cuda' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Wan22 Test"
+        source_file_dependencies:
+          - tests/e2e/offline_inference/test_wan22_t2v.py
+          - tests/e2e/online_serving/test_wan22_t2v.py
+          - vllm_omni/diffusion/models/wan2_2/
+        commands:
+          - |
+            timeout 40m bash -c "
+              export VLLM_IMAGE_FETCH_TIMEOUT=60
+              pytest -s -v tests/e2e/offline_inference/test_wan22_t2v.py tests/e2e/online_serving/test_wan22_t2v.py -m 'advanced_model and cuda and diffusion' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · HunyuanVideo 1.5 Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_hunyuan_video_15.py
+          - vllm_omni/diffusion/models/hunyuan_video/
+        timeout_in_minutes: 40
+        commands:
+          - |
+            timeout 40m bash -c "
+              pytest -s -v tests/e2e/online_serving/test_hunyuan_video_15.py -m 'advanced_model and cuda and diffusion' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Z Image Test"
+        source_file_dependencies:
+          - tests/e2e/offline_inference/test_z_image.py
+          - tests/e2e/online_serving/test_z_image.py
+          - vllm_omni/diffusion/models/z_image/
+        timeout_in_minutes: 30
+        commands:
+          - pytest -s -v tests/e2e/offline_inference/test_z_image.py tests/e2e/online_serving/test_z_image.py -m 'advanced_model and cuda and diffusion' --run-level 'advanced_model'
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · FLUX.2 Klein Test"
+        source_file_dependencies:
+          - tests/e2e/offline_inference/test_flux2_klein.py
+          - tests/e2e/online_serving/test_flux2_klein.py
+          - vllm_omni/diffusion/models/flux2_klein/
+        timeout_in_minutes: 30
+        commands:
+          - pytest -s -v tests/e2e/offline_inference/test_flux2_klein.py tests/e2e/online_serving/test_flux2_klein.py -m 'advanced_model and cuda and diffusion' --run-level 'advanced_model'
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · SD3 Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_sd3.py
+          - vllm_omni/diffusion/models/sd3/
+        timeout_in_minutes: 45
+        commands:
+          - |
+            timeout 45m bash -c "
+              export VLLM_IMAGE_FETCH_TIMEOUT=60
+              pytest -s -v tests/e2e/online_serving/test_sd3.py -m 'advanced_model and cuda and diffusion' --run-level 'advanced_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 4
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate

.buildkite/test-npu-nightly.yaml

@@ -0,0 +1,24 @@
+steps:
+  - group: ":card_index_dividers: Diffusion X2V Model Test"
+    key: nightly-diffusion-x2v-group
+    depends_on: upload-nightly-pipeline
+    if: >-
+      build.env("NIGHTLY") == "1" ||
+      build.pull_request.labels includes "nightly-test" ||
+      build.pull_request.labels includes "diffusion-x2v-test"
+    steps:
+      - label: ":full_moon: Diffusion X2V · Function Test"
+        agents:
+          queue: "ascend-a2b3"
+          resource_class: "npu-4"
+        image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}"
+        plugins:
+          - kubernetes:
+              podSpecPatch:
+                imagePullSecrets:
+                  - name: swr-secret
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          HF_TOKEN: "${HF_TOKEN}"
+        commands:
+          - pytest -s -v tests/e2e/online_serving/test_wan22_expansion.py -k npu_i2v --run-level "full_model"

.buildkite/test-npu-ready.yaml

@@ -0,0 +1,21 @@
+steps:
+  - group: ":card_index_dividers: TTS Model Test"
+    key: ready-tts-group
+    depends_on: upload-ready-pipeline
+    steps:
+      - label: ":sound: TTS · Qwen3-TTS CustomVoice Test"
+        agents:
+          queue: "ascend-a2b3"
+          resource_class: "npu-1"
+        image: "swr.cn-southwest-2.myhuaweicloud.com/modelfoundry/vllm-omni-ci-npu:${BUILDKITE_COMMIT}"
+        plugins:
+          - kubernetes:
+              podSpecPatch:
+                imagePullSecrets:
+                  - name: swr-secret
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
+          HF_TOKEN: "${HF_TOKEN}"
+        commands:
+          - pytest -s -v tests/e2e/online_serving/test_qwen3_tts_customvoice.py -m 'core_model' --run-level 'core_model'

.buildkite/test-ready.yml

@@ -0,0 +1,674 @@
+env:
+  VLLM_WORKER_MULTIPROC_METHOD: spawn
+  VLLM_USE_FLASHINFER_MOE_FP16: 0
+  HF_HUB_DOWNLOAD_TIMEOUT: 300
+  HF_HUB_ETAG_TIMEOUT: 60
+
+steps:
+  - group: ":card_index_dividers: Simple Test"
+    depends_on: upload-ready-pipeline
+    steps:
+      - label: "Simple · Diffusion & Model Executor Test"
+        commands:
+          - "timeout 20m pytest -sv tests/diffusion tests/model_executor -m 'core_model and cpu' --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Simple · Other Test"
+        commands:
+          - "timeout 20m pytest -sv -m 'core_model and cpu' --ignore=tests/diffusion --ignore=tests/model_executor --cov=vllm_omni --cov-report=term-missing:skip-covered --cov-report=xml"
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+  - label: "Custom Pipeline Test"
+    depends_on: upload-ready-pipeline
+    commands:
+      - timeout 20m pytest -s -v tests/e2e/offline_inference/custom_pipeline/ -m "core_model"
+    agents:
+      queue: "gpu_1_queue"
+    plugins:
+      - docker#v5.2.0:
+          image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+          always-pull: true
+          propagate-environment: true
+          environment:
+            - "HF_HOME=/fsx/hf_cache"
+          volumes:
+            - "/fsx/hf_cache:/fsx/hf_cache"
+
+  - group: ":card_index_dividers: Diffusion Test"
+    depends_on: upload-ready-pipeline
+    steps:
+      - label: "Diffusion · Offloader Test"
+        commands:
+          - timeout 40m pytest -s -v tests/diffusion/offloader -m "core_model and cuda" --run-level "core_model"
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · Cache Test"
+        commands:
+          - timeout 15m pytest -sv tests/diffusion/cache -m 'core_model and cuda' --run-level "core_model"
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · Batch Test"
+        commands:
+          - timeout 15m pytest -sv tests/diffusion/batching -m 'core_model and cuda' --run-level "core_model"
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · Distributed Test"
+        commands:
+          - timeout 20m pytest -sv tests/diffusion/distributed -m 'core_model and cuda' --run-level "core_model"
+        agents:
+          queue: "gpu_4_queue" # g6.12xlarge instance on AWS, has 4 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · Model Test"
+        commands:
+          - timeout 15m pytest -sv tests/diffusion/models/ -m 'core_model and cuda' --run-level "core_model"
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · Other Test"
+        commands:
+          - timeout 15m pytest -sv tests/diffusion/*.py -m 'core_model and cuda' --run-level "core_model"
+        agents:
+          queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+
+  - label: "Engine&Model Executor Test"
+    depends_on: upload-ready-pipeline
+    commands:
+      - |
+        timeout 15m bash -c "
+                  pytest -sv tests/engine/ tests/model_executor/ -m 'core_model and cuda' --run-level 'core_model'
+        "
+    agents:
+      queue: "gpu_1_queue" # g6.12xlarge instance on AWS, has 4 L4 GPU
+    plugins:
+      - docker#v5.2.0:
+          image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+          always-pull: true
+          propagate-environment: true
+          shm-size: "8gb"
+          environment:
+            - "HF_HOME=/fsx/hf_cache"
+            - "HF_TOKEN"
+          volumes:
+            - "/fsx/hf_cache:/fsx/hf_cache"
+
+
+  - label: "Distributed Test"
+    depends_on: upload-ready-pipeline
+    commands:
+      - |
+        timeout 20m bash -c "
+                  pytest -sv tests/distributed/ -m 'core_model and cuda' --run-level 'core_model'
+        "
+    agents:
+      queue: "mithril-h100-pool"
+    plugins:
+      - kubernetes:
+          podSpec:
+            containers:
+              - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                resources:
+                  limits:
+                    nvidia.com/gpu: 1
+                volumeMounts:
+                  - name: devshm
+                    mountPath: /dev/shm
+                  - name: hf-cache
+                    mountPath: /root/.cache/huggingface
+                env:
+                  - name: HF_HOME
+                    value: /root/.cache/huggingface
+            nodeSelector:
+              node.kubernetes.io/instance-type: gpu-h100-sxm
+            volumes:
+              - name: devshm
+                emptyDir:
+                  medium: Memory
+              - name: hf-cache
+                hostPath:
+                  path: /mnt/hf-cache
+                  type: DirectoryOrCreate
+
+  - group: ":card_index_dividers: Entrypoints Test"
+    depends_on: upload-ready-pipeline
+    steps:
+      - label: "Entrypoints Test with H100"
+        commands:
+          - |
+            timeout 30m bash -c "
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              pytest -sv tests/entrypoints/ -m 'core_model and cuda and H100' --run-level 'core_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 2
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Entrypoints Test with L4"
+        commands:
+          - |
+            timeout 30m bash -c "
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              pytest -sv tests/entrypoints/ -m 'core_model and cuda and L4' --run-level 'core_model'
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+  - group: ":card_index_dividers: E2E Test"
+    depends_on: upload-ready-pipeline
+    steps:
+      - label: "RLHF · VeRL-Omni E2E Test"
+        source_file_dependencies:
+          - tests/e2e/offline_inference/rlhf_test/test_verl_omni_e2e.py
+          - tests/e2e/offline_inference/custom_pipeline/
+          - vllm_omni/diffusion/
+          - vllm_omni/engine/
+          - vllm_omni/entrypoints/
+          - vllm_omni/inputs/
+          - vllm_omni/outputs/
+        commands:
+          - |
+            timeout 20m bash -c "
+              pytest -s -v tests/e2e/offline_inference/rlhf_test/test_verl_omni_e2e.py
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Omni · Qwen3-Omni Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen3_omni.py
+          - vllm_omni/model_executor/models/qwen3_omni/
+          - vllm_omni/model_executor/stage_input_processors/qwen3_omni.py
+          - vllm_omni/deploy/qwen3_omni_moe.yaml
+        commands:
+          - |
+            timeout 20m bash -c "
+              pytest -s -v tests/e2e/online_serving/test_qwen3_omni.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 2
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "TTS · VoxCPM2 Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_voxcpm2_tts.py
+          - vllm_omni/model_executor/models/voxcpm2/
+          - vllm_omni/deploy/voxcpm2.yaml
+        timeout_in_minutes: 30
+        commands:
+          - |
+            timeout 30m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              pytest -s -v tests/e2e/online_serving/test_voxcpm2_tts.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "TTS · Qwen3-TTS CustomVoice Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen3_tts_customvoice.py
+          - vllm_omni/model_executor/models/qwen3_tts/
+          - vllm_omni/model_executor/stage_input_processors/qwen3_tts.py
+          - vllm_omni/deploy/qwen3_tts.yaml
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
+              pytest -s -v tests/e2e/online_serving/test_qwen3_tts_customvoice.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "TTS · Qwen3-TTS Base Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen3_tts_base.py
+          - vllm_omni/model_executor/models/qwen3_tts/
+          - vllm_omni/model_executor/stage_input_processors/qwen3_tts.py
+          - vllm_omni/deploy/qwen3_tts.yaml
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
+              export VLLM_OMNI_USE_V2_RUNNER=1
+              pytest -s -v tests/e2e/online_serving/test_qwen3_tts_base.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "TTS · Higgs-Audio-V3 Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_higgs_audio_v3.py
+          - vllm_omni/model_executor/models/higgs_audio_v3/
+          - vllm_omni/model_executor/stage_input_processors/higgs_audio_v3.py
+          - vllm_omni/deploy/higgs_multimodal_qwen3.yaml
+        commands:
+          - |
+            timeout 20m bash -c "
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              export VLLM_USE_DEEP_GEMM=0
+              export VLLM_MOE_USE_DEEP_GEMM=0
+              pytest -s -v tests/e2e/online_serving/test_higgs_audio_v3.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"
+
+      - label: "Diffusion · Qwen Image Edit Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen_image_edit.py
+          - vllm_omni/diffusion/models/qwen_image/
+        commands:
+          - |
+            timeout 20m bash -c "
+              pytest -s -v tests/e2e/online_serving/test_qwen_image_edit.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Bagel Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_bagel.py
+          - vllm_omni/model_executor/models/bagel/
+          - vllm_omni/diffusion/models/bagel/
+          - vllm_omni/model_executor/stage_input_processors/bagel.py
+          - vllm_omni/deploy/bagel.yaml
+        commands:
+          - |
+            timeout 40m bash -c "
+              export VLLM_IMAGE_FETCH_TIMEOUT=60
+              pytest -s -v tests/e2e/online_serving/test_bagel.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Wan22 Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_wan22_t2v.py
+          - vllm_omni/diffusion/models/wan2_2/
+        commands:
+          - |
+            timeout 40m bash -c "
+              export VLLM_IMAGE_FETCH_TIMEOUT=60
+              pytest -s -v tests/e2e/online_serving/test_wan22_t2v.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+
+      - label: "Diffusion · Qwen Image Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen_image.py
+          - vllm_omni/diffusion/models/qwen_image/
+        commands:
+          - |
+            timeout 40m bash -c "
+              export VLLM_IMAGE_FETCH_TIMEOUT=60
+              pytest -s -v tests/e2e/online_serving/test_qwen_image.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      - label: "Diffusion · Qwen Image Layered Test"
+        source_file_dependencies:
+          - tests/e2e/online_serving/test_qwen_image_layered.py
+          - vllm_omni/diffusion/models/qwen_image/
+        commands:
+          - |
+            timeout 40m bash -c "
+              export VLLM_IMAGE_FETCH_TIMEOUT=60
+              pytest -s -v tests/e2e/online_serving/test_qwen_image_layered.py -m 'core_model' --run-level 'core_model'
+            "
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 1
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate

.buildkite/test-template-amd-omni.j2

@@ -57,7 +57,7 @@ steps:
           {% endif %}
 {% set cmd_body = (step.command or (step.commands | join("\n"))) | trim %}
 {% set indented_cmd = cmd_body | replace("\n", "\n            ") %}
-        command: bash .buildkite/amd/scripts/run-amd-test.sh
+        command: bash .buildkite/scripts/hardware_ci/run-amd-test.sh
         env:
           DOCKER_BUILDKIT: "1"
 {% if step.env %}
@@ -104,7 +104,7 @@ steps:
       {% endif %}
 {% set cmd_body = (step.command or (step.commands | join("\n"))) | trim %}
 {% set indented_cmd = cmd_body | replace("\n", "\n        ") %}
-    command: bash .buildkite/amd/scripts/run-amd-test.sh
+    command: bash .buildkite/scripts/hardware_ci/run-amd-test.sh
     env:
       DOCKER_BUILDKIT: "1"
 {% if step.env %}

.buildkite/test-weekly.yml

@@ -0,0 +1,148 @@
+env:
+  VLLM_WORKER_MULTIPROC_METHOD: spawn
+  VLLM_USE_FLASHINFER_MOE_FP16: 0
+  HF_HUB_DOWNLOAD_TIMEOUT: 300
+  HF_HUB_ETAG_TIMEOUT: 60
+
+steps:
+  - label: "Reliability Test - qwen3-omni"
+    timeout_in_minutes: 180
+    depends_on: upload-weekly-pipeline
+    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
+    commands:
+      - pytest -s -v tests/dfx/reliability/test_reliability_qwen3_omni.py -m "slow"
+    agents:
+      queue: "mithril-h100-pool"
+    plugins:
+      - kubernetes:
+          podSpec:
+            containers:
+              - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                resources:
+                  limits:
+                    nvidia.com/gpu: 2
+                volumeMounts:
+                  - name: devshm
+                    mountPath: /dev/shm
+                  - name: hf-cache
+                    mountPath: /root/.cache/huggingface
+                env:
+                  - name: HF_HOME
+                    value: /root/.cache/huggingface
+                  - name: HF_TOKEN
+                    valueFrom:
+                      secretKeyRef:
+                        name: hf-token-secret
+                        key: token
+            nodeSelector:
+              node.kubernetes.io/instance-type: gpu-h100-sxm
+            volumes:
+              - name: devshm
+                emptyDir:
+                  medium: Memory
+              - name: hf-cache
+                hostPath:
+                  path: /mnt/hf-cache
+                  type: DirectoryOrCreate
+
+  - label: "Reliability Test - wan22"
+    timeout_in_minutes: 180
+    depends_on: upload-weekly-pipeline
+    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
+    commands:
+      - pytest -s -v tests/dfx/reliability/test_reliability_wan22.py -m "slow"
+    agents:
+      queue: "mithril-h100-pool"
+    plugins:
+      - kubernetes:
+          podSpec:
+            containers:
+              - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                resources:
+                  limits:
+                    nvidia.com/gpu: 2
+                volumeMounts:
+                  - name: devshm
+                    mountPath: /dev/shm
+                  - name: hf-cache
+                    mountPath: /root/.cache/huggingface
+                env:
+                  - name: HF_HOME
+                    value: /root/.cache/huggingface
+                  - name: HF_TOKEN
+                    valueFrom:
+                      secretKeyRef:
+                        name: hf-token-secret
+                        key: token
+            nodeSelector:
+              node.kubernetes.io/instance-type: gpu-h100-sxm
+            volumes:
+              - name: devshm
+                emptyDir:
+                  medium: Memory
+              - name: hf-cache
+                hostPath:
+                  path: /mnt/hf-cache
+                  type: DirectoryOrCreate
+
+  # Invalid HTTP / WS payloads vs live Omni servers (Issue #3649 tracks known skips).
+  - group: "Reliability Test - Invalid parameters Test"
+    depends_on: upload-weekly-pipeline
+    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
+    steps:
+      - label: "Invalid parameters Test · H100"
+        timeout_in_minutes: 120
+        commands:
+          - pytest -s -v tests/dfx/reliability/invalid_param_test/ -m "slow and H100"
+        agents:
+          queue: "mithril-h100-pool"
+        plugins:
+          - kubernetes:
+              podSpec:
+                containers:
+                  - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+                    resources:
+                      limits:
+                        nvidia.com/gpu: 2
+                    volumeMounts:
+                      - name: devshm
+                        mountPath: /dev/shm
+                      - name: hf-cache
+                        mountPath: /root/.cache/huggingface
+                    env:
+                      - name: HF_HOME
+                        value: /root/.cache/huggingface
+                      - name: HF_TOKEN
+                        valueFrom:
+                          secretKeyRef:
+                            name: hf-token-secret
+                            key: token
+                nodeSelector:
+                  node.kubernetes.io/instance-type: gpu-h100-sxm
+                volumes:
+                  - name: devshm
+                    emptyDir:
+                      medium: Memory
+                  - name: hf-cache
+                    hostPath:
+                      path: /mnt/hf-cache
+                      type: DirectoryOrCreate
+
+      # Stable Audio Open, Qwen3-TTS speech/batch/voices paths marked ``L4``.
+      - label: "Invalid parameters Test · L4"
+        timeout_in_minutes: 120
+        commands:
+          - pytest -s -v tests/dfx/reliability/invalid_param_test/ -m "slow and L4"
+        agents:
+          queue: "gpu_1_queue"
+        plugins:
+          - docker#v5.2.0:
+              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+              always-pull: true
+              propagate-environment: true
+              shm-size: "8gb"
+              environment:
+                - "HF_HOME=/fsx/hf_cache"
+                - "HF_TOKEN"
+              volumes:
+                - "/fsx/hf_cache:/fsx/hf_cache"

.claude/skills/add-diffusion-model/SKILL.md

@@ -24,7 +24,7 @@ Check the model's HF repo for `model_index.json`. This determines your path:
 
 | Scenario | How to identify | Migration path |
 |----------|----------------|----------------|
-| **Already supported** | `_class_name` in `model_index.json` matches a key in `_DIFFUSION_MODELS` in `registry.py` | Skip implementation, then validate model-specific examples, tests, and docs as needed |
+| **Already supported** | `_class_name` in `model_index.json` matches a key in `_DIFFUSION_MODELS` in `registry.py` | Skip to Step 5 (test) and Step 7 (docs) |
 | **Diffusers-based** | Has standard `model_index.json` with `_diffusers_version`, subfolders for `transformer/`, `vae/`, etc. | Follow **Path A** below |
 | **Custom/private repo** | No diffusers `model_index.json`, weights in non-standard format, custom model code in a separate git repo | Follow **Path B** below |
 | **Hybrid** | Has some diffusers components (VAE) but custom transformer/fusion | Mix of Path A and Path B |
@@ -52,7 +52,7 @@ vllm_omni/diffusion/models/your_model_name/
 2. Replace attention with `vllm_omni.diffusion.attention.layer.Attention` (QKV shape: `[B, seq, heads, head_dim]`).
 3. Add `od_config: OmniDiffusionConfig | None = None` to `__init__`.
 4. Add `load_weights()` method mapping diffusers weight names to vllm-omni names.
-5. Add class attributes for acceleration features such as `_repeated_blocks` and `_layerwise_offload_blocks_attrs` (see [references/transformer-adaptation.md](references/transformer-adaptation.md) for examples).
+5. Add class attributes: `_repeated_blocks`, `_layerwise_offload_blocks_attr`.
 
 ### A4. Adapt pipeline
 
@@ -78,34 +78,13 @@ class YourPipeline(nn.Module):
 
 Add post/pre-process functions in the same pipeline file. Register them in `registry.py`.
 
-### A4.1 Add progress bar support (recommended)
-
-For pipelines with a standard denoising loop, prefer the existing progress bar pattern instead of hand-rolled logging.
-
-```python
-from vllm_omni.diffusion.models.progress_bar import ProgressBarMixin
-
-class YourPipeline(nn.Module, ProgressBarMixin):
-    def forward(self, req: OmniDiffusionRequest) -> DiffusionOutput:
-        # ... prepare timesteps / latents ...
-        with self.progress_bar(total=len(timesteps)) as progress_bar:
-            for i, t in enumerate(timesteps):
-                # predict noise / scheduler step
-                latents = ...
-                progress_bar.update()
-
-        return DiffusionOutput(output=output)
-```
-
-For custom loop structures, follow `vllm_omni/diffusion/models/progress_bar.py` and existing pipelines using `ProgressBarMixin`.
-
 ### A5. Register, test, docs → continue at Step 4 below.
 
 ---
 
 ## Path B: Custom/Private Repo Model
 
-For models without a diffusers pipeline — weights in custom format, model code in a private repo. Real examples: DreamID-Omni, BAGEL, HunyuanImage3. See [references/custom-model-patterns.md](references/custom-model-patterns.md) for concrete custom-model integration patterns.
+For models without a diffusers pipeline — weights in custom format, model code in a private repo. Real examples: DreamID-Omni, BAGEL, HunyuanImage3.
 
 ### B1. Understand the reference repo
 
@@ -155,7 +134,7 @@ Copy the essential model files into `vllm_omni/diffusion/models/<name>/` and ada
 
 ### B4. Handle custom weight loading
 
-Custom models have two common patterns for weight loading:
+Custom models have two patterns for weight loading:
 
 **Pattern 1: Bypass standard loader** (DreamID-Omni style)
 
@@ -170,10 +149,8 @@ class CustomPipeline(nn.Module):
         self.vae = custom_init_vae(model, device=self.device)
         self.text_encoder = custom_init_text_encoder(model, device=self.device)
         self.transformer = CustomFusionModel(CONFIG)
-        load_custom_checkpoint(
-            self.transformer,
-            checkpoint_path=os.path.join(model, "model.safetensors"),
-        )
+        load_custom_checkpoint(self.transformer,
+            checkpoint_path=os.path.join(model, "model.safetensors"))
         # NO weights_sources defined — bypasses standard loader
 
     def load_weights(self, weights):
@@ -185,8 +162,6 @@ class CustomPipeline(nn.Module):
 When weights are in safetensors format but need name remapping:
 
 ```python
-from vllm.model_executor.model_loader.weight_utils import default_weight_loader
-
 class CustomPipeline(nn.Module):
     def __init__(self, *, od_config, prefix=""):
         super().__init__()
@@ -212,9 +187,7 @@ class CustomPipeline(nn.Module):
             # Remap original weight names to vllm-omni module names
             name = self._remap_weight_name(name)
             if name in params:
-                param = params[name]
-                weight_loader = getattr(param, "weight_loader", default_weight_loader)
-                weight_loader(param, tensor)
+                default_weight_loader(params[name], tensor)
                 loaded.add(name)
         return loaded
 ```
@@ -308,16 +281,16 @@ For Omni or custom models, create:
 ### Step 7: Update Documentation
 
 Required updates:
-1. `docs/user_guide/diffusion/parallelism/overview.md` — parallelism support overview/table
-2. `docs/user_guide/diffusion/cpu_offload.md` — if CPU offload supported (add to supported models table)
-3. `docs/user_guide/diffusion/cache_acceleration/teacache.md` — if TeaCache supported
-4. `docs/user_guide/diffusion/cache_acceleration/cache_dit.md` — if Cache-DiT supported
-5. Offline example docs under `examples/offline_inference/<name>/` (`README.md` or category-specific `.md`)
-6. `examples/online_serving/<name>/README.md` — online serving docs
+1. `docs/user_guide/diffusion/parallelism_acceleration.md` — parallelism support table
+2. `docs/user_guide/diffusion/cpu_offload_diffusion.md` — if CPU offload supported (add to supported models table)
+3. `docs/user_guide/diffusion/teacache.md` — if TeaCache supported
+4. `docs/user_guide/diffusion/cache_dit_acceleration.md` — if Cache-DiT supported
+5. `examples/offline_inference/xxx/README.md` — offline example docs
+6. `examples/online_serve/xxx/README.md` — online serve docs
 
 ### Step 8: Add E2E Tests
 
-**Follow the [vllm-omni-test skill](../vllm-omni-test/SKILL.md)** for markers, file naming, Buildkite wiring, and run commands. Also read [l4_functionality_tests.inc.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/test_examples/l4_functionality_tests.inc.md), [test_system_overview.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/test_system_overview.md), and [test_writing_guide.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/test_writing_guide.md).
+**Follow the [vllm-omni-test skill](../vllm-omni-test/SKILL.md)** for markers, file naming, Buildkite wiring, and run commands. Also read [l4_functionality_tests.inc.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/test_examples/l4_functionality_tests.inc.md) and [CI_5levels.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/CI_5levels.md).
 
 Classify the model's **CI priority** first:
 
@@ -334,7 +307,7 @@ Classify the model's **CI priority** first:
 | **L1** | `tests/diffusion/models/{slug}/`, `tests/diffusion/cache/`, transformer unit tests | `core_model` + `cpu` | `test-ready.yml` | Weight remap, `_sp_plan`, cache enabler registration, shape contracts |
 | **L2** | `tests/e2e/online_serving/test_{slug}.py` (and offline if the category is offline-first) | **`core_model` + `advanced_model`** (both on baseline smoke) + `diffusion` + `@hardware_test` / `hardware_marks` | `test-ready.yml` | Default deploy smoke — minimal `num_inference_steps`, single prompt |
 | **L3** | `tests/e2e/online_serving/test_{slug}.py` **and** `tests/e2e/offline_inference/test_{slug}.py` when offline matters | Baseline smoke: **`core_model` + `advanced_model`**; heavier cases: `advanced_model` only (+ `diffusion`) | `test-merge.yml` **or** merged into nightly diffusion function job | Real weights, streaming/API paths, LoRA/offload smoke |
-| **L4** | `tests/e2e/online_serving/test_{slug}_expansion.py` (+ offline expansion if needed) | `full_model` + `diffusion` | `test-nightly.yml` (X2I / X2V / X2A function groups) | Feature combos per [#1832](https://github.com/vllm-project/vllm-omni/issues/1832); perf → `tests/dfx/perf/tests/test_{model}_vllm_omni.json` with per-case **`mark`** (`hardware_marks` + `full_model` + `diffusion`) |
+| **L4** | `tests/e2e/online_serving/test_{slug}_expansion.py` (+ offline expansion if needed) | `full_model` + `diffusion` | `test-nightly.yml` (X2I / X2V / X2A function groups) | Feature combos per [#1832](https://github.com/vllm-project/vllm-omni/issues/1832); perf → `tests/dfx/perf/tests/test_{model}_vllm_omni.json` |
 
 **L2 & L3 online — same file, dual marks on the baseline smoke:** The **first / simplest** case in `test_{slug}.py` (default deploy, minimal steps, single prompt) should carry **both** `@pytest.mark.core_model` and `@pytest.mark.advanced_model` on the **same** function so L2 (`test-ready.yml`) and L3 (`test-merge.yml`) share one smoke test. Heavier deploy variants or API paths in the same file use **`advanced_model` only**. When L3 moves to nightly, migrate those heavier cases into `test_{slug}_expansion.py` with `full_model` and remove the dedicated `test-merge.yml` job (see `test_longcat_image_expansion.py`, `test_qwen_image_expansion.py`).
 
@@ -418,8 +391,7 @@ def enable_cache_for_your_model(pipeline, cache_config):
 
 #### 9d. Register the custom enabler
 
-Add your enabler to `CUSTOM_DIT_ENABLERS` in
-`vllm_omni/diffusion/cache/cachedit/model_specific.py`:
+Add your enabler to `CUSTOM_DIT_ENABLERS` in `vllm_omni/diffusion/cache/cache_dit_backend.py`:
 
 ```python
 CUSTOM_DIT_ENABLERS = {
@@ -565,7 +537,7 @@ class YourTransformer(nn.Module):
 #### 10e. Update parallelism documentation
 
 After adding parallelism support, update:
-1. `docs/user_guide/diffusion/parallelism/overview.md` — add your model to the support overview/table
+1. `docs/user_guide/diffusion/parallelism_acceleration.md` — add your model to the support table
 2. Record which parallelism methods are supported (USP, Ring, CFG, TP, HSDP, VAE-Patch)
 
 ### Step 11: Add CPU Offload Support

.claude/skills/add-diffusion-model/references/cache-dit-patterns.md

@@ -29,13 +29,8 @@ vLLM-Omni integrates cache-dit through `CacheDiTBackend`:
 | `cache_dit.refresh_context()` | Updates cache context when `num_inference_steps` changes |
 
 **Source files**:
-- `vllm_omni/diffusion/cache/cachedit/__init__.py` — public package API
-- `vllm_omni/diffusion/cache/cachedit/backend.py` — generic backend lifecycle and integration
-- `vllm_omni/diffusion/cache/cachedit/config.py` — Cache-DiT configuration types
-- `vllm_omni/diffusion/cache/cachedit/model_specific.py` — model-specific adapters, enablers, and registry
-
-Outside `vllm_omni/diffusion/cache`, import Cache-DiT symbols only from
-`vllm_omni.diffusion.cache.cachedit`.
+- `vllm_omni/diffusion/cache/cache_dit_backend.py` — `CacheDiTBackend`, enablers, `CUSTOM_DIT_ENABLERS`
+- `vllm_omni/diffusion/cache/` — cache backend implementations
 
 ## Standard Models: Automatic Support
 
@@ -180,8 +175,7 @@ Inspect your block's `forward()` return type and residual connection pattern to
 
 ## Registering Custom Enablers
 
-Add your enabler to `CUSTOM_DIT_ENABLERS` in
-`vllm_omni/diffusion/cache/cachedit/model_specific.py`:
+Add your enabler to `CUSTOM_DIT_ENABLERS` in `vllm_omni/diffusion/cache/cache_dit_backend.py`:
 
 ```python
 CUSTOM_DIT_ENABLERS = {
@@ -254,7 +248,7 @@ Models listed in `_NO_CACHE_ACCELERATION` in `vllm_omni/diffusion/registry.py` d
 
 | Model | Path | Notes |
 |-------|------|-------|
-| Standard DiT | `cachedit.backend::enable_cache_for_dit` | Default enabler, automatic |
-| Wan2.2 | `cachedit.model_specific::enable_cache_for_wan22` | Dual-transformer, auto-detects mode |
-| LongCat | `cachedit.config::CacheDiTAdapterConfig` | Declarative multi-block-list adapter |
-| BAGEL | `cachedit.model_specific::BagelCachedAdapter` | Custom cached adapter |
+| Standard DiT | `cache_dit_backend.py::enable_cache_for_dit` | Default enabler, automatic |
+| Wan2.2 | `cache_dit_backend.py::enable_cache_for_wan22` | Dual-transformer, auto-detects mode |
+| LongCat | `cache_dit_backend.py::enable_cache_for_longcat_image` | Multi-block-list |
+| BAGEL | `cache_dit_backend.py::enable_cache_for_bagel` | Complex omni model |

.claude/skills/add-diffusion-model/references/parallelism-patterns.md

@@ -569,4 +569,4 @@ Check which parallelism methods a model supports:
 | **TP** | Uses `ColumnParallelLinear` / `QKVParallelLinear`. Search: `grep -r 'ParallelLinear\|QKVParallel' vllm_omni/diffusion/models/<model>/` |
 | **HSDP** | Transformer defines `_hsdp_shard_conditions`. Search: `grep -r '_hsdp_shard_conditions' vllm_omni/diffusion/models/` |
 
-The canonical per-model support overview/table is in `docs/user_guide/diffusion/parallelism/overview.md`.
+The canonical per-model support table is in `docs/user_guide/diffusion/parallelism_acceleration.md`.

.claude/skills/add-diffusion-model/references/transformer-adaptation.md

@@ -219,14 +219,13 @@ def load_weights(self, weights):
 
 ### Class-Level Attributes for Features
 
-
 ```python
 class MyTransformer(nn.Module):
     # torch.compile: list block class names that repeat and can be compiled
     _repeated_blocks = ["MyTransformerBlock"]
 
     # CPU offload: attribute name of the nn.ModuleList containing blocks
-    _layerwise_offload_blocks_attrs = ["blocks"]
+    _layerwise_offload_blocks_attr = "blocks"
 
     # LoRA: mapping of fused param names to original param names
     packed_modules_mapping = {"to_qkv": ["to_q", "to_k", "to_v"]}

.claude/skills/add-tts-model/SKILL.md

@@ -307,7 +307,7 @@ def build_voice_clone_prompt(ref_audio_path: str, text: str, codec) -> list:
 
 ### Test Case Writing (CI Levels)
 
-**Follow the [vllm-omni-test skill](../vllm-omni-test/SKILL.md)** for markers, file naming (`test_{slug}.py` / `test_{slug}_expansion.py`), Buildkite wiring, and copy-paste run commands. Also read [test_system_overview.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/test_system_overview.md) and [test_writing_guide.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/test_writing_guide.md).
+**Follow the [vllm-omni-test skill](../vllm-omni-test/SKILL.md)** for markers, file naming (`test_{slug}.py` / `test_{slug}_expansion.py`), Buildkite wiring, and copy-paste run commands. Also read [CI_5levels.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/CI_5levels.md) and [tests_style.md](https://github.com/vllm-project/vllm-omni/blob/main/docs/contributing/ci/tests_style.md).
 
 Classify the model's **CI priority** first (high / medium / low). High-priority TTS models are typically those on the integration hot path or listed in tracking issues such as [#1832](https://github.com/vllm-project/vllm-omni/issues/1832); medium and low tiers cover the long tail. When unsure, ask the reviewer which tier applies.
 

.claude/skills/precheck-pr/SKILL.md

@@ -63,8 +63,6 @@ If multiple rows apply (e.g., a diffusion model is also a new model), union the
 
 Ask: "Quick mode or full mode?" Then walk the checklist for the detected PR type from [references/checklists.md](references/checklists.md). Each item produces ✓, ✗, or ⚠.
 
-**Also run the Code-Quality sweep on every PR**, regardless of type or mode: the five diff-scoped checks in [references/code-quality.md](references/code-quality.md) — kwargs fragility, broad-except swallow, `Any`/wrong type hints, hot-path `.clone()`/`deepcopy`, and event-loop blocking — plus the advisory conventions (log level, structured logging, synchronization, cleanup, dependencies, naming) in [checklists.md](references/checklists.md). These count only lines the PR *adds* — the pre-existing backlog across the repo is out of scope.
-
 ### Step 5: Print Report
 
 ```
@@ -76,7 +74,6 @@ Pre-check report for <branch>
   Dimension          Result
   ─────────────────  ──────
   PR title format    ✓
-  Code quality       ⚠ 1 broad except, 2 Any hints
   PR desc integrity  ✓
   Registry/config    ✓
   Dead code          ⚠ 2 warnings

.claude/skills/precheck-pr/references/checklists.md

@@ -35,31 +35,6 @@ PR title must follow the project convention documented in `docs/contributing/REA
 
 ---
 
-## All PRs: Code Quality (run on every PR, quick & full)
-
-Diff-scoped sweep for five fragility patterns that are pervasive in the existing codebase — **only lines the PR adds count**, not the pre-existing backlog. Detection commands, codebase examples, and full severity rules live in [code-quality.md](code-quality.md).
-
-- [ ] **No new `**kwargs` string-lookup plumbing:** the diff adds no `**kwargs` + `kwargs["..."]` / `kwargs.get("...")` / `"..." in kwargs`. ⚠ any new instance; ✗ if a new string key is duplicated across ≥2 files without a shared constant, or unknown keys are silently dropped on a fail-fast path.
-- [ ] **No new broad exception swallow:** the diff adds no `except Exception:` / bare `except:`. ⚠ any new instance; ✗ for `except: pass` / `except Exception: return None` / `continue` on a fail-fast path (init, config validation, weight loading, request handling).
-- [ ] **No new `Any` / wrong type hints:** the diff adds no `: Any` / `-> Any` / untyped production signature, and no new `SimpleNamespace` in tests faking an object that already has a real typed stub. ⚠ any new instance; ✗ for a *wrong* (actively misleading) annotation or a `SimpleNamespace` test fake of a typed object.
-- [ ] **No new hot-path copy:** the diff adds no `.clone()` / `.copy_()` / `copy.deepcopy(...)` inside a per-step loop or per-request path. ⚠ any new instance; ✗ for `.clone()` of a full latent/activations tensor inside an AR or diffusion step loop without a comment, or `deepcopy` of a scheduler / sampling-config on the request path.
-- [ ] **No new event-loop blocking:** the diff adds no `time.sleep` / blocking HTTP (`requests`, `urllib`) / `await` inside a held lock in `async def` code. ⚠ any new instance; ✗ for blocking sleep or HTTP on the serving path (`engine/`, `entrypoints/`, async `connectors/`), or a lock held across an `await`. Blocking inside a dedicated worker thread is fine.
-
-Roll the ⚠/✗ counts into the report as a single **Code quality** dimension row.
-
-### Conventions (eyeball — no grep)
-
-Judgment calls that don't grep cleanly — apply them while reading the diff, not as a mechanical check:
-
-- **Right log level** — `info` is for end-user-visible events; routine internal events are `debug`; hot paths are `trace` or removed. Logging is not free.
-- **Structured fields over f-strings** — pass dynamic values as structured fields (`logger.info("msg", extra={...})`), not interpolated strings, so they stay greppable and parseable.
-- **No reflexive synchronization** — don't add `threading.Lock` / `asyncio.Lock` / `deepcopy` / re-wrap shared state unless there is real concurrent mutation; the GIL covers most Python code. Owners decide their own synchronization — don't pre-lock in a constructor.
-- **Context managers for cleanup** — use `with` / `contextlib` for files, locks, CUDA streams, and profilers; not raw `open()` + manual `.close()` or ad-hoc `finally`.
-- **No new dependencies without justification** — prefer the stdlib and existing vllm-omni deps; a new top-level dep needs a reason (check `pyproject.toml` / `requirements*.txt`).
-- **Naming / interface discipline** — don't rename or change interfaces just for taste in this PR, but do fix names that actively mislead (e.g. `serve` implying a long-running server, `Instance` where multiple instance kinds coexist).
-
----
-
 ## Bug Fix PRs
 
 ### Quick
@@ -74,7 +49,7 @@ Judgment calls that don't grep cleanly — apply them while reading the diff, no
 - [ ] **Regression test exists:** `git diff --name-only` includes at least one `tests/` file
 - [ ] **Regression test reproduces the original error:** the test would fail on main, pass on this branch
 - [ ] **Fix matches root cause exactly:** no "fix the symptom + something else" — if the root cause is dtype, the fix is dtype, not dtype + reformatting
-- [ ] **No silent failure risk:** no bare `except: pass`, no `try: ... except: return None`, no empty fallback added. See the broad-except pattern in [code-quality.md](code-quality.md#2-broad-exception-swallow) for the full severity rule (⚠ any new broad catch; ✗ for a swallow on a fail-fast path).
+- [ ] **No silent failure risk:** no bare `except: pass`, no `try: ... except: return None`, no empty fallback added
 - [ ] **Upstream pattern match:** the fix follows the same pattern as existing code for similar cases (grep for analogous `from_pretrained` calls, etc.)
 - [ ] **Environment documented:** torch/transformers/vllm versions listed if the bug is version-dependent
 

.claude/skills/precheck-pr/references/code-quality.md

@@ -1,179 +0,0 @@
-# Code-Quality Patterns
-
-Five fragility patterns to sweep on **every** PR, in both quick and full mode. These are pervasive in the existing codebase (hundreds of pre-existing sites) — so the checks are **diff-scoped**: only lines the PR *adds* count. Whole-repo greps light up every PR with the backlog and are useless.
-
-All commands assume:
-```bash
-BASE="$(git merge-base HEAD origin/main)"
-```
-
-**Severity:** ⚠ for any *new* instance introduced by the diff; ✗ for the acute sub-cases defined per pattern. Acute ✗ findings belong in the report's blocking count.
-
----
-
-## 1. `**kwargs` string-lookup passthrough
-
-### Anti-pattern
-```python
-def forward(self, **kwargs):
-    # BLOCKER-prone: raw dict + string keys duplicated across files
-    if "runtime_additional_information" in kwargs and \
-       "model_intermediate_buffer" not in kwargs:
-        ...
-```
-Real instances of this exact guard are copy-pasted across `cosyvoice3.py`, `indextts2_talker.py` (×2), `indextts2_s2mel_decoder.py`, `qwen3_tts_talker.py`, `qwen3_omni.py`. A typo in any string key silently no-ops; there is no type check, no IDE completion, and unknown keys are silently dropped.
-
-### Why dangerous
-Raw `dict` + `**kwargs` + lists of param-name strings is **fragile by construction**: the compiler can't see a missing or misspelled key, so failures are silent. Each duplicated copy is a new place for the strings to drift.
-
-### Severity
-- ⚠ any *new* `**kwargs` signature whose body does `kwargs["..."]` / `kwargs.get("...")` / `"..." in kwargs`.
-- ✗ when the diff introduces a **new** string key that is duplicated across ≥2 files without a shared module constant, or when `**kwargs` plumbing **silently drops unknown keys** on a fail-fast path (init, config validation, weight loading).
-
-### Detect (diff-scoped)
-```bash
-git diff ${BASE}...HEAD -- "*.py" | grep "^+[^+]" \
-  | grep -E '\*\*kwargs|kwargs(\.get|\[)|"[a-z_]+"\s+in\s+kwargs'
-```
-
-### Fix
-Prefer explicit typed params. The repo already has the right primitive: `msgspec.Struct` with `forbid_unknown_fields=True` rejects unknown keys instead of silently dropping them — see `_StructBase` in `vllm_omni/data_entry_keys.py:105` and `MoriPullRequest` in `vllm_omni/distributed/omni_connectors/connectors/mori_transfer_engine_connector.py:81`. A `@dataclass` or `TypedDict` works too.
-
-If `**kwargs` is unavoidable (vLLM base-class compat), centralize the key strings as module-level constants behind **one** typed accessor, instead of repeating the literal across files.
-
----
-
-## 2. Broad exception swallow
-
-### Anti-pattern
-```python
-# BLOCKER-prone: catches everything, returns None on any failure
-try:
-    evt = build_event(...)
-    return evt
-except Exception:
-    return None
-```
-Real clusters: `vllm_omni/metrics/stats.py:240,274,367,450` and `vllm_omni/metrics/{modality,utils}.py` — all `except Exception: return None`.
-
-### Why dangerous
-A generic `except Exception:` is **too flexible**. If we catch an exception type we don't expect, it is almost always an unhandled edge case that needs to be *fixed* — not swallowed into `None`/`pass`/`continue`. Broad catches turn fixable bugs into silent wrong behavior and make debugging impossible.
-
-The repo backs this with ruff `BLE001` (broad-exception catch); it is currently suppressed via `# noqa: BLE001` in `vllm_omni/patch.py:413`. Prefer fixing over suppressing.
-
-### Severity
-- ⚠ any *new* `except Exception:` / `except BaseException:` / bare `except:`.
-- ✗ for `except:` / `except Exception: pass|return None|continue` on a **fail-fast path** (init, config validation, weight loading, request handling, connector setup). This extends the Bug-Fix "No silent failure risk" ✗ in [checklists.md](checklists.md).
-
-### Detect (diff-scoped)
-```bash
-# Any new broad catch (⚠ baseline):
-git diff ${BASE}...HEAD -- "*.py" | grep "^+[^+]" \
-  | grep -E 'except\s*(Exception|BaseException)?\s*:'
-# Note: this matches only the immediate-colon swallow forms (`except :`,
-# `except Exception:`) — it deliberately excludes specific catches like
-# `except ValueError:` and the `except Exception as e:` variant. Eyeball
-# the latter; `as e` usually means it is logged, but still verify.
-```
-
-### Fix
-Catch the **specific** types you actually expect (`ValueError`, `KeyError`, `AttributeError`, `OSError`, `torch.cuda.OutOfMemoryError`, `msgspec.ValidationError`, …). At a genuine top-level / best-effort boundary (metrics, signal handlers), at minimum **log** the exception; never swallow into a bare `pass`/`return None` on a path that should fail loudly.
-
----
-
-## 3. `Any` / wrong type annotations
-
-### Anti-pattern
-```python
-# ⚠ Any param + Any return — no contract, no narrowing
-def _extract_mm_output(engine_outputs: Any) -> dict[str, Any]:
-    ...
-```
-
-### Why dangerous (root cause)
-`Any` disables the type checker for that position, so typos and wrong shapes pass silently. The repo has 636 `: Any` params, 244 `-> Any` returns, and **1,173 `SimpleNamespace` in `tests/`**. The `Any` leak is largely test-driven: vibe-coded unit tests fake objects with `SimpleNamespace`, which forces the production code consuming them to be typed `Any`, and the `Any` then propagates outward into public signatures.
-
-### Severity
-- ⚠ any *new* `: Any` / `-> Any` / fully-untyped signature in production code.
-- ✗ for a **wrong** annotation (actively misleading — e.g. `-> bool` that returns `Optional[bool]`), or a *new* `SimpleNamespace` in a test that mimics an object which already has a real typed stub / `@dataclass` / `TypedDict` / `Protocol`.
-
-### Detect (diff-scoped)
-```bash
-# Any leaking into new signatures:
-git diff ${BASE}...HEAD -- "*.py" | grep "^+[^+]" \
-  | grep -E ':\s*Any\b|->\s*Any\b'
-# SimpleNamespace newly added in tests (the Any-leak source):
-git diff ${BASE}...HEAD -- "tests/" | grep "^+[^+]" | grep "SimpleNamespace"
-```
-
-### Fix
-Replace `Any` with the concrete type, a `Protocol`, or a `Union`. If a type is genuinely dynamic, prefer `object` + `isinstance` narrowing over `Any` — `object` keeps the checker engaged. In tests, replace `SimpleNamespace` with the real class, a small `@dataclass`/`TypedDict`, or a `Protocol` + fake, so the system under test stays fully typed.
-
----
-
-## 4. Unnecessary copy — `.clone()` / `copy.deepcopy` in hot paths
-
-### Anti-pattern
-```python
-# Per-step in a diffusion/AR loop: clones the full latent every iteration
-for t in timesteps:
-    trajectory_latents.append(x_t.clone())   # O(N) GPU alloc + memcpy each step
-```
-Real clusters: `vllm_omni/diffusion/models/bagel/bagel_transformer.py` calls `x_t.clone()` ~9× across a trajectory loop (lines 1820–2186); the cache hooks `vllm_omni/diffusion/cache/teacache/hook.py:155` and `magcache/hook.py:182` clone `hidden_states` per step. Separately, `copy.deepcopy(self.scheduler)` / `copy.deepcopy(req.sampling_params)` is copy-pasted across diffusion pipelines — `pipeline_qwen_image.py:786`, `pipeline_hunyuan_image3.py:1904`, `pipeline_ltx2*.py` (×3), `pipeline_dreamzero.py:1085`, `diffusion_model_runner.py:398` — a full object-graph deepcopy on the request path. (Contrast `dreamzero/causal_wan_model.py:203`, which carries a `# No .clone() needed.` comment — the aware pattern.)
-
-### Why dangerous
-A `.clone()` of a GPU tensor or a `copy.deepcopy` of a config object inside a per-step / per-request path is a hidden allocator: it reads as one Python line but triggers a GPU memcpy or a deep object-graph walk every iteration. Across 50 diffusion steps or hundreds of AR steps it dominates latency and VRAM bandwidth. `deepcopy` additionally ignores any custom device / `__copy__` logic and can drag tensors through CPU.
-
-### Severity
-- ⚠ any *new* `.clone()` / `.copy_()` / `copy.deepcopy(...)` inside a per-step loop or per-request path.
-- ✗ for `.clone()` of a full latent/activations tensor inside an AR or diffusion step loop without an explanatory comment, or `copy.deepcopy` of a scheduler / sampling-config on the request hot path.
-
-### Detect (diff-scoped)
-```bash
-git diff ${BASE}...HEAD -- "*.py" | grep "^+[^+]" \
-  | grep -E '\.clone\(\)|\.copy_\(|copy\.deepcopy|deepcopy\('
-```
-
-### Fix
-Tensors: prefer a view (`x[..., :k]`, `torch.narrow`, `as_strided`) or write into a pre-allocated buffer with `dst.copy_(src)` instead of `.clone()`; if the caller no longer needs the original, `move`/`del` it rather than copy. Schedulers / configs: build a fresh lightweight instance per request, or split out the mutable per-request bits so the shared immutable part isn't deepcopied. If a copy is genuinely required (e.g. the source is mutated in place later), keep it and add a one-line comment saying why.
-
----
-
-## 5. Blocking the asyncio event loop / lock held across `await`
-
-### Anti-pattern
-```python
-# BLOCKER-prone: blocking sleep / blocking HTTP / lock held across an await,
-# all inside an async (event-loop) function
-async def handle(self, req):
-    with self._lock:
-        resp = await self._client.call(req)   # lock held across await -> serializes all requests
-    time.sleep(0.1)                           # stalls the single loop for every concurrent request
-```
-
-### Why dangerous
-vLLM-Omni serves over a single asyncio event loop. Any synchronous blocking work on an `async def` path — `time.sleep`, blocking HTTP (`requests`/`urllib`), heavy CPU work, or a lock held across an `await` — stalls the loop for *every* concurrent request, not just the blocking one. A lock held across `await` silently serializes the whole pipeline; blocking I/O tanks throughput and inflates tail latency.
-
-### Severity
-- ⚠ any *new* `time.sleep(`, blocking HTTP call (`requests.get/post`, `urllib.request`, `urlopen`), or `await` inside a `with Lock` / `async with` critical section, in `async def` code.
-- ✗ for `time.sleep` or blocking HTTP on the serving path (`engine/`, `entrypoints/`, async `connectors/`), or a lock acquired and held across an `await` on the request path.
-
-### Detect (diff-scoped)
-```bash
-# New blocking primitives introduced by the diff — then eyeball whether each
-# sits in an async function on the loop vs a dedicated worker thread:
-git diff ${BASE}...HEAD -- "*.py" | grep "^+[^+]" \
-  | grep -E 'time\.sleep|requests\.(get|post|put|delete)|urllib\.request|urlopen'
-# Lock acquired with an await potentially inside the critical section (manual eyeball):
-git diff ${BASE}...HEAD -- "*.py" | grep "^+[^+]" | grep -E 'async with|\.acquire\(\)'
-```
-Note: blocking inside a **dedicated worker thread is correct** — the repo already does this (`diffusion/model_loader/hub_prefetch.py`, `distributed/omni_connectors/kv_transfer_manager.py`, the `mooncake`/`mori` connectors poll with `time.sleep` on their own threads). The check targets the *event-loop* path only.
-
-### Fix
-Move blocking work off the loop with `asyncio.to_thread(...)` / `run_in_executor`, or use a native async client (`httpx.AsyncClient` / `aiohttp`). Never `await` while holding a lock — drop the lock before the `await` and re-acquire after, or restructure so the critical section contains no `await`.
-
----
-
-## Running the sweep
-
-Run all five detections against `${BASE}...HEAD`, then for each hit decide ⚠ vs ✗ using the severity rules above. Roll the results into the report as a single **Code quality** dimension row (count of ⚠ and ✗), alongside the type-specific checklist from [checklists.md](checklists.md).

.claude/skills/vllm-omni-test/SKILL.md

@@ -9,7 +9,7 @@ description: Generate and run tests for vllm-project/vllm-omni with CI-aligned l
 
 Use this skill to generate minimal, stable test cases and run them with the correct marker/level strategy for [vllm-project/vllm-omni](https://github.com/vllm-project/vllm-omni).
 
-**Link convention:** Paths such as `.buildkite/` and `docs/contributing/` live at the **vllm-omni repo root**. Markdown links use repo-relative paths from this skill file (e.g. `../../../.buildkite/cuda/test-ready.yml`, `../../../docs/contributing/ci/test_system_overview.md`).
+**Link convention:** Paths such as `.buildkite/` and `docs/contributing/` live at the **vllm-omni repo root**. Markdown links use repo-relative paths from this skill file (e.g. `../../../.buildkite/test-ready.yml`, `../../../docs/contributing/ci/CI_5levels.md`).
 
 Default priorities:
 
@@ -96,8 +96,8 @@ Existing references: `tests/e2e/offline_inference/test_qwen2_5_omni.py` (L2-styl
 |----------|------------------|---------------------------|---------------------------|
 | **Offline inference e2e** | `tests/e2e/offline_inference/` | **Module (default):** `omni_runner` + `omni_runner_handler`. **Function (isolation only):** `omni_runner_function` + `omni_runner_handler_function`. Diffusion/TTS may use `Omni(...).generate` directly | L2: `core_model` + **one of** `omni` / `tts` / `diffusion`; `@hardware_test(...)` when GPU/NPU is required |
 | **Online serving e2e** | `tests/e2e/online_serving/` | **Module (default):** `omni_server` + `openai_client`. **Function (isolation only):** `omni_server_function` + `openai_client_function`. Clients: `send_omni_request` (omni), `send_audio_speech_request` (tts), `send_diffusion_request` / `send_video_diffusion_request` / `send_images_generations_request` (diffusion) | Baseline smoke: **`core_model` + `advanced_model`**; heavier paths: `advanced_model` only; L4 expansion: `full_model` |
-| **Documentation / runnable examples** | `tests/examples/offline_inference/`, `tests/examples/online_serving/` | **Offline docs (preferred):** extract Python/Bash blocks from the doc README (e.g. `ReadmeSnippet.extract_readme_snippets`), `pytest.mark.parametrize` each snippet, run via `example_runner.run` with a stable `output_subfolder`. **Online docs:** copy client/request scripts into dedicated tests and keep them in sync with the doc page. | Usually **L4**: `advanced_model`, often `example` plus hardware marks matching the nightly docs-example job (see `.buildkite/cuda/test-nightly.yml`). Full conventions: [docs/contributing/ci/test_examples/l4_doc_example_tests.inc.md](../../../docs/contributing/ci/test_examples/l4_doc_example_tests.inc.md) (introduced in [PR #1910](https://github.com/vllm-project/vllm-omni/pull/1910): naming, output directory layout, skip rules, avoid trimming `num_inference_steps` without a strong CI reason). |
-| **Performance / benchmark** | `tests/dfx/perf/tests/*.json` + `run_*_benchmark.py` | JSON or script-driven server + load config; assert explicit metrics / baselines | L4 Perf: JSON `mark` with `full_model` + `omni`/`tts`/`diffusion`; wire `test-nightly.yml` Perf steps |
+| **Documentation / runnable examples** | `tests/examples/offline_inference/`, `tests/examples/online_serving/` | **Offline docs (preferred):** extract Python/Bash blocks from the doc README (e.g. `ReadmeSnippet.extract_readme_snippets`), `pytest.mark.parametrize` each snippet, run via `example_runner.run` with a stable `output_subfolder`. **Online docs:** copy client/request scripts into dedicated tests and keep them in sync with the doc page. | Usually **L4**: `advanced_model`, often `example` plus hardware marks matching the nightly docs-example job (see `.buildkite/test-nightly.yml`). Full conventions: [docs/contributing/ci/test_examples/l4_doc_example_tests.inc.md](../../../docs/contributing/ci/test_examples/l4_doc_example_tests.inc.md) (introduced in [PR #1910](https://github.com/vllm-project/vllm-omni/pull/1910): naming, output directory layout, skip rules, avoid trimming `num_inference_steps` without a strong CI reason). |
+| **Performance / benchmark** | `tests/dfx/perf/tests/*.json` + `run_*_benchmark.py` | JSON or script-driven server + load config; assert explicit metrics / baselines | L4 Perf: `full_model` + `benchmark`; wire `test-nightly.yml` Perf steps |
 | **Invalid parameter / negative HTTP validation** | `tests/dfx/reliability/invalid_param_test/` | Live `omni_server` + low-level `send_*_http_request` with `err_code` / `err_message` | `pytest.mark.slow` + `omni` / `tts` / `diffusion` + `@hardware_marks` (`H100` or `L4`); CI in **`test-weekly.yml`** (not ready/merge/nightly) |
 
 **If the user’s test plan includes invalid parameter validation / invalid params / negative HTTP / 400 validation:** do **not** add those `test_*` functions to `tests/e2e/online_serving/test_*.py` or `*_expansion.py`. **Move or author them** under `tests/dfx/reliability/invalid_param_test/` in the **endpoint-matching script** (see **Invalid parameter validation** below). Success-path e2e and invalid-param dfx tests must stay in separate modules.
@@ -270,7 +270,7 @@ def test_text_to_video_001(omni_server, openai_client) -> None:
 
 *Documentation example tests:* follow the **Preferred Test Strategy** in [l4_doc_example_tests.inc.md](../../../docs/contributing/ci/test_examples/l4_doc_example_tests.inc.md): dynamic extraction for offline READMEs; explicit copied client code for online pages until extraction is justified; use the documented **naming**, **output directory** (page folder + case id), and **skipping** rules (e.g. Gradio-only scripts).
 
-*Performance tests:* add or extend entries under `tests/dfx/perf/tests/` (and JSON configs where the project uses them), with **explicit baselines**, **`mark`** on each case (`hardware_marks` + `full_model` + type marker), and the same nightly Perf step pattern as in-tree configs.
+*Performance tests:* add or extend entries under `tests/perf/` (and JSON configs where the project uses them), with **explicit baselines** and the same marker/run-level pairing as the CI step that will execute them.
 
 **3. Cross-cutting rules**
 
@@ -460,7 +460,7 @@ When the user asks for **invalid parameter validation**, **invalid request bodie
 
 4. **Adding a new model** to invalid-param coverage: append a `pytest.param(OmniServerParams(model="...", stage_config_path=..., server_args=...), ...)` entry to the script’s `_PARAMS` list — do **not** create `test_invalid_<model>.py` unless the route is new.
 
-5. **CI — `.buildkite/cuda/test-weekly.yml` only** (not `test-ready.yml` / `test-merge.yml` / `test-nightly.yml`):
+5. **CI — `.buildkite/test-weekly.yml` only** (not `test-ready.yml` / `test-merge.yml` / `test-nightly.yml`):
 
 | Weekly step | Command | When your cases run |
 |-------------|---------|---------------------|
@@ -559,7 +559,7 @@ L1 tests under `tests/entrypoints/` may keep **minimal** asserts next to the han
 
 #### Omni Test Writing Guidance (L1-L4 Layering)
 
-When the goal is a general Omni/multimodal test case, prioritize mapping the test to the correct purpose, directory, and resource assumptions aligned with the layers (see [test_system_overview.md](../../../docs/contributing/ci/test_system_overview.md) and [test_writing_guide.md](../../../docs/contributing/ci/test_writing_guide.md)):
+When the goal is a general Omni/multimodal test case, prioritize mapping the test to the correct purpose, directory, and resource assumptions aligned with the layers (see [CI_5levels.md](../../../docs/contributing/ci/CI_5levels.md)):
 
 - **L1**: Unit/logic validation on CPU (`core_model and cpu`). Cover input validation, branches, and exception paths (`tests/<component>/test_*.py`). **Mock with `mocker` / `monkeypatch` only — not `unittest.mock`.**
 - **L2**: Basic e2e (online/offline basic scenarios). Prefer dummy/lightweight models to validate the end-to-end request-to-output-structure/streaming chain (typically `tests/e2e/online_serving/` and `tests/e2e/offline_inference/`).
@@ -592,31 +592,27 @@ Implementation strategy:
 **When the user explicitly asks for L4 perf (or full L4 including perf):**
 
 1. **Do not** put throughput/latency assertions inside `test_*_expansion.py` — perf uses the **dfx benchmark harness**, not `omni_server` e2e fixtures.
-2. Add or extend a **JSON config** under `tests/dfx/perf/tests/` (mirror in-tree names: `test_qwen_image_vllm_omni.json`, `test_cosmos3_vllm_omni.json`, `test_qwen3_omni_async_chunk.json`).
-3. Each JSON **case**: `test_name`, optional **`mark`** (`hardware_marks` required when present; `marks`: `full_model` + `omni`/`tts`/`diffusion`), `server_params` (`model`, `serve_args` or `stage_config_name`), `benchmark_params[]` with **`name`** and explicit **`baseline`** (`throughput_qps`, `latency_mean`, `peak_memory_mb_mean`, …). Diffusion cases also set `server_type` (e.g. `"vllm-omni"`) and usually `benchmark_endpoint`.
+2. Add or extend a **JSON config** under `tests/dfx/perf/tests/` (mirror in-tree names: `test_qwen_image_vllm_omni.json`, `test_bagel_vllm_omni.json`).
+3. Each JSON entry: `test_name`, `server_params` (`model`, `serve_args`), `benchmark_params[]` with explicit **`baseline`** (`throughput_qps`, `latency_mean`, `peak_memory_mb_mean`, …).
 4. Run locally via the matching script (model type → runner):
 
 | Model type | Runner script | Example config |
 |------------|---------------|----------------|
 | **Diffusion X2I/X2V** | `tests/dfx/perf/scripts/run_diffusion_benchmark.py` | `tests/dfx/perf/tests/test_qwen_image_vllm_omni.json` |
 | **TTS** | `tests/dfx/perf/scripts/run_benchmark.py` | `tests/dfx/perf/tests/test_tts.json` (shared) or `test_{slug}.json` (dedicated; use when the model must not join the shared nightly matrix before integration — e.g. `test_voxcpm2.json`, `test_coqui_tts.json`) |
-| **Omni** | `tests/dfx/perf/scripts/run_benchmark.py` | `test_qwen3_omni_no_async_chunk.json`, `test_qwen3_omni_async_chunk.json`, `test_qwen3_omni_vllm_text.json`, `test_qwen3_omni_multi_replicas.json` |
+| **Omni** | (see existing Omni Perf steps — `run_benchmark.py` / dedicated omni configs in-tree) | `tests/dfx/perf/tests/test_omni*.json` |
 
-5. Wire **`test-nightly.yml` · Perf Test** step for that model (separate from Function Test): export `DIFFUSION_BENCHMARK_DIR` / `BENCHMARK_DIR`, run pytest on the script with **`--test-config-file`** (nightly perf steps do not use `-m`), **upload artifacts** (`buildkite-agent artifact upload`), often **multi-GPU H100** for diffusion perf.
+5. Wire **`test-nightly.yml` · Perf Test** step for that model (separate from Function Test): export `DIFFUSION_BENCHMARK_DIR` / `BENCHMARK_DIR`, run pytest on the script with `--test-config-file`, **upload artifacts** (`buildkite-agent artifact upload`), often **multi-GPU H100** for diffusion perf.
 6. **One benchmark scenario → one `test_name` block** in JSON (same spirit as one `test_*` per function case). Combine server `serve_args` + workload in one entry; use multiple `benchmark_params` rows for size/step sweeps under the same server config.
 
-**Perf local commands:**
+**Perf local command (diffusion example):**
 
 ```bash
 cd tests
 export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
 export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
-# Single file (CI-like)
 pytest -s -v dfx/perf/scripts/run_diffusion_benchmark.py \
   --test-config-file dfx/perf/tests/test_<model>_vllm_omni.json
-# Bulk load + filter by JSON mark
-pytest -sv dfx/perf/scripts/run_diffusion_benchmark.py -m "full_model and diffusion and H100"
-pytest -sv dfx/perf/scripts/run_benchmark.py -m "full_model and omni and H100"
 ```
 
 **Clarify in the test plan** which L4 pillars you deliver: `Function only` | `Function + Perf` | `Function + Accuracy + Perf`.
@@ -629,10 +625,10 @@ If the test is not already collected by an existing pipeline command (for exampl
 
 | Test level | Edit this file | Typical trigger / intent |
 |------------|----------------|---------------------------|
-| **L1** and **L2** | [`.buildkite/cuda/test-ready.yml`](../../../.buildkite/cuda/test-ready.yml) | PR **ready** label; L1 CPU + L2 GPU/basic e2e (steps labeled **Omni ·**, **TTS ·**, **Diffusion ·**) |
-| **L3** | [`.buildkite/cuda/test-merge.yml`](../../../.buildkite/cuda/test-merge.yml) | Post-merge; `advanced_model` integration per model type |
-| **L4** | [`.buildkite/cuda/test-nightly.yml`](../../../.buildkite/cuda/test-nightly.yml) | Nightly; grouped by model type (see below) |
-| **Invalid param / reliability (weekly)** | [`.buildkite/cuda/test-weekly.yml`](../../../.buildkite/cuda/test-weekly.yml) | Weekly; `tests/dfx/reliability/invalid_param_test/` — **not** L1–L4 e2e pipelines |
+| **L1** and **L2** | [`.buildkite/test-ready.yml`](../../../.buildkite/test-ready.yml) | PR **ready** label; L1 CPU + L2 GPU/basic e2e (steps labeled **Omni ·**, **TTS ·**, **Diffusion ·**) |
+| **L3** | [`.buildkite/test-merge.yml`](../../../.buildkite/test-merge.yml) | Post-merge; `advanced_model` integration per model type |
+| **L4** | [`.buildkite/test-nightly.yml`](../../../.buildkite/test-nightly.yml) | Nightly; grouped by model type (see below) |
+| **Invalid param / reliability (weekly)** | [`.buildkite/test-weekly.yml`](../../../.buildkite/test-weekly.yml) | Weekly; `tests/dfx/reliability/invalid_param_test/` — **not** L1–L4 e2e pipelines |
 
 **Level-specific Buildkite delivery (follow the user’s requested level):**
 
@@ -809,21 +805,21 @@ Use **repo-relative paths** (no leading `./`). Prefer **directory** entries (`..
 
 When extending an **existing** E2E step to run an additional test file, **append** that file (and any new model/deploy paths) to `source_file_dependencies` on the same step. When authoring a PR, mention the dependency list in the **Buildkite change** section of your test plan.
 
-#### L3 / L4: validating Buildkite from a feature branch (`cuda/pipeline.yml`)
+#### L3 / L4: validating Buildkite from a feature branch (root `pipeline.yml`)
 
-Production behavior is driven by [`.buildkite/cuda/pipeline.yml`](../../../.buildkite/cuda/pipeline.yml): it builds the CI image, then uploads **L2** (`test-ready.yml` on non-`main`), **L3** (`test-merge.yml` when `build.branch == "main"`), and **L4** (`test-nightly.yml` when `build.env("NIGHTLY") == "1"`).
+Production behavior is driven by [`.buildkite/pipeline.yml`](../../../.buildkite/pipeline.yml): it builds the CI image, then uploads **L2** (`test-ready.yml` on non-`main`), **L3** (`test-merge.yml` when `build.branch == "main"`), and **L4** (`test-nightly.yml` when `build.env("NIGHTLY") == "1"`).
 
 When you add or debug **L3** or **L4** tests and need those child pipelines to run **off `main`** or **without** `NIGHTLY=1`, apply **temporary** edits (revert before merge):
 
 1. **Point the upload at the right file (optional shortcut)**  
-   In the **Upload Ready Pipeline** step, the command is `buildkite-agent pipeline upload .buildkite/cuda/test-ready.yml`. For a one-off run you may change that path to `.buildkite/cuda/test-merge.yml` (L3) or `.buildkite/cuda/test-nightly.yml` (L4), and adjust the step’s `if:` so it does not conflict with the other upload steps (avoid double-uploading unless intentional).
+   In the **Upload Ready Pipeline** step, the command is `buildkite-agent pipeline upload .buildkite/test-ready.yml`. For a one-off run you may change that path to `.buildkite/test-merge.yml` (L3) or `.buildkite/test-nightly.yml` (L4), and adjust the step’s `if:` so it does not conflict with the other upload steps (avoid double-uploading unless intentional).
 
 2. **L3 — merge pipeline**  
-   The gate `if: build.branch == "main"` lives on the **Upload Merge Pipeline** step in `cuda/pipeline.yml` (not inside `test-merge.yml`). **Comment out that `if` line** so `test-merge.yml` is uploaded on your feature branch. Alternatively rely on step (1) instead of the dedicated merge upload step.
+   The gate `if: build.branch == "main"` lives on the **Upload Merge Pipeline** step in `pipeline.yml` (not inside `test-merge.yml`). **Comment out that `if` line** so `test-merge.yml` is uploaded on your feature branch. Alternatively rely on step (1) instead of the dedicated merge upload step.
 
 3. **L4 — nightly pipeline**  
-   **Comment out** `if: build.env("NIGHTLY") == "1"` on the **Upload Nightly Pipeline** step in `cuda/pipeline.yml` so the nightly definition is uploaded without setting the env var.  
-   Child steps in [`test-nightly.yml`](../../../.buildkite/cuda/test-nightly.yml) often repeat `if: build.env("NIGHTLY") == "1"`; **comment out those lines on the steps you need to run**, otherwise they will still be skipped after upload.
+   **Comment out** `if: build.env("NIGHTLY") == "1"` on the **Upload Nightly Pipeline** step in `pipeline.yml` so the nightly definition is uploaded without setting the env var.  
+   Child steps in [`test-nightly.yml`](../../../.buildkite/test-nightly.yml) often repeat `if: build.env("NIGHTLY") == "1"`; **comment out those lines on the steps you need to run**, otherwise they will still be skipped after upload.
 
 ### Step 6: Run Tests
 
@@ -890,5 +886,5 @@ When completing a request, return:
 ## Additional Resources
 
 - Marker and command routing: [references/test-routing.md](references/test-routing.md)
-- CI pipelines (vllm-omni): [test-ready.yml](../../../.buildkite/cuda/test-ready.yml) (L1/L2), [test-merge.yml](../../../.buildkite/cuda/test-merge.yml) (L3), [test-nightly.yml](../../../.buildkite/cuda/test-nightly.yml) (L4), [test-weekly.yml](../../../.buildkite/cuda/test-weekly.yml) (invalid param / reliability)
+- CI pipelines (vllm-omni): [test-ready.yml](../../../.buildkite/test-ready.yml) (L1/L2), [test-merge.yml](../../../.buildkite/test-merge.yml) (L3), [test-nightly.yml](../../../.buildkite/test-nightly.yml) (L4), [test-weekly.yml](../../../.buildkite/test-weekly.yml) (invalid param / reliability)
 - L4 documentation example tests (naming, extraction vs copied scripts, output dirs, skips): [docs/contributing/ci/test_examples/l4_doc_example_tests.inc.md](../../../docs/contributing/ci/test_examples/l4_doc_example_tests.inc.md) — see also [PR #1910](https://github.com/vllm-project/vllm-omni/pull/1910)

.claude/skills/vllm-omni-test/references/test-routing.md

@@ -2,7 +2,7 @@
 
 Use this reference to map testing goals to levels, markers, and runnable commands.
 
-**Repo paths** (`.buildkite/`, `docs/contributing/…`): link with repo-relative paths from this file (e.g. `../../../../.buildkite/cuda/test-ready.yml`, `../../../../docs/contributing/ci/test_system_overview.md`).
+**Repo paths** (`.buildkite/`, `docs/contributing/…`): link with repo-relative paths from this file (e.g. `../../../../.buildkite/test-ready.yml`, `../../../../docs/contributing/ci/CI_5levels.md`).
 
 ## Model-centric e2e filename convention (L2–L4)
 
@@ -127,12 +127,8 @@ pytest -s -v e2e/online_serving/test_qwen_image_expansion.py -m "full_model and
 cd tests
 export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
 export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
-# CI-like: single JSON file (nightly Perf Test steps)
 pytest -s -v dfx/perf/scripts/run_diffusion_benchmark.py \
   --test-config-file dfx/perf/tests/test_qwen_image_vllm_omni.json
-# Local bulk load: all *.json under tests/dfx/perf/tests/, filter by JSON mark
-pytest -sv dfx/perf/scripts/run_diffusion_benchmark.py -m "full_model and diffusion and H100"
-pytest -sv dfx/perf/scripts/run_benchmark.py -m "full_model and omni and H100"
 ```
 
 Broad marker sweep (when no explicit file shard):
@@ -154,26 +150,7 @@ Nightly **L4** for a model is often **multiple jobs** in `test-nightly.yml`, not
 
 **Rule:** “L4 functional cases” → **Function only** by default. State that Perf/Accuracy are separate pillars; add them only when requested.
 
-**Diffusion perf JSON** (`tests/dfx/perf/tests/test_<slug>_vllm_omni.json`): array of **case objects**, each with:
-
-| Field | Required | Notes |
-|-------|----------|-------|
-| `test_name` | Yes | Unique server/workload id |
-| `mark` | Recommended | When present: **`hardware_marks` required** + optional `marks` (`full_model`, `diffusion`, `local_model`, …). Parsed by `resolve_pytest_marks`; enables local `-m` filtering. |
-| `server_type` | Diffusion | e.g. `"vllm-omni"` — routes case to `run_diffusion_benchmark.py` |
-| `benchmark_endpoint` | Diffusion | e.g. `/v1/videos`, `/v1/images/generations` |
-| `server_params` | Yes | `model`, `serve_args`, … |
-| `benchmark_params[]` | Yes | Each row: **`name`** (pytest id suffix), workload fields, **`baseline`** |
-
-Copy structure from `test_qwen_image_vllm_omni.json` or `test_cosmos3_vllm_omni.json` (2-GPU mark example).
-
-**Omni perf JSON**: `test_qwen3_omni_*.json` — same `mark` shape with `"marks": ["full_model", "omni"]`; run via `run_benchmark.py`.
-
-**Runner split**: `is_diffusion_perf_config()` → diffusion when `server_type` is set or `"diffusion"` ∈ `mark.marks`; otherwise omni/tts → `run_benchmark.py`.
-
-**Param ids**: `{test_name}-{benchmark_params.name}` (e.g. `test_tts-p0`, not `test_tts-0`).
-
-**Result filenames**: runtime hardware from `get_runtime_resource_label()`; `H100` omitted on default CI pool.
+**Diffusion perf JSON** (`tests/dfx/perf/tests/test_<slug>_vllm_omni.json`): array of objects with `test_name`, `server_params` (`model`, `serve_args`), `benchmark_params[]`, and per-workload **`baseline`** (`throughput_qps`, `latency_mean`, `peak_memory_mb_mean`). Copy structure from `test_qwen_image_vllm_omni.json` in-tree.
 
 **Diffusion perf nightly step** (under **Diffusion X2I(&A&T)**; copy full `kubernetes` plugins from `· Perf Test · Qwen-Image`):
 
@@ -200,9 +177,9 @@ Copy structure from `test_qwen_image_vllm_omni.json` or `test_cosmos3_vllm_omni.
 
 | Model type | Perf runner | Config example |
 |------------|-------------|----------------|
-| Diffusion X2I/X2V | `dfx/perf/scripts/run_diffusion_benchmark.py` | `test_qwen_image_vllm_omni.json`, `test_cosmos3_vllm_omni.json` |
-| TTS | `dfx/perf/scripts/run_benchmark.py` | `test_tts.json`, `test_voxcpm2.json`, `test_higgs_audio_v3.json` |
-| Omni | `dfx/perf/scripts/run_benchmark.py` | `test_qwen3_omni_async_chunk.json`, `test_qwen3_omni_no_async_chunk.json`, … |
+| Diffusion X2I/X2V | `dfx/perf/scripts/run_diffusion_benchmark.py` | `test_qwen_image_vllm_omni.json` |
+| TTS | `dfx/perf/scripts/run_benchmark.py` | `test_tts.json` (`BENCHMARK_DIR`) |
+| Omni | `run_benchmark.py` / omni configs in-tree | `test_omni*.json` |
 
 Do **not** put throughput/latency baselines inside `test_*_expansion.py` — that belongs in the dfx perf JSON + nightly Perf job.
 
@@ -224,7 +201,7 @@ When the user’s test plan includes **invalid parameter validation**, **invalid
 | Sleep / wakeup / server control | `test_invalid_server_control.py` |
 
 3. **Style:** `pytestmark = [pytest.mark.slow, pytest.mark.<type>]`; `_PARAMS` + `hardware_marks`; `send_*_http_request` with `err_code` + `err_message`; parametrized `body_spec` rows with `id=`; `_minimal_*_json()` helpers; `_SKIP_ISSUE_3649` when tracked in [#3649](https://github.com/vllm-project/vllm-omni/issues/3649).
-4. **CI:** [`.buildkite/cuda/test-weekly.yml`](../../../../.buildkite/cuda/test-weekly.yml) group **Reliability Test - Invalid parameters Test** — **not** ready/merge/nightly.
+4. **CI:** [`.buildkite/test-weekly.yml`](../../../../.buildkite/test-weekly.yml) group **Reliability Test - Invalid parameters Test** — **not** ready/merge/nightly.
 
 ```bash
 # Weekly H100 (diffusion / omni / video invalid-param)
@@ -291,10 +268,10 @@ When adding or modifying tests, do not stop at “where the file lives” — al
 
 | Level | Repo file | Model-type grouping |
 |-------|-----------|---------------------|
-| L1, L2 | [`.buildkite/cuda/test-ready.yml`](../../../../.buildkite/cuda/test-ready.yml) | Steps prefixed **Omni ·**, **TTS ·**, **Diffusion ·** under **E2E Test** — **`source_file_dependencies` required** |
-| L3 | [`.buildkite/cuda/test-merge.yml`](../../../../.buildkite/cuda/test-merge.yml) | Per-model E2E steps; `-m "advanced_model and …"` — **`source_file_dependencies` required** |
-| L4 | [`.buildkite/cuda/test-nightly.yml`](../../../../.buildkite/cuda/test-nightly.yml) | **Omni / TTS / Diffusion X2I(&A&T) / X2V** — each group may have **Function**, **Accuracy**, **Perf**, **Doc** steps |
-| Invalid param (weekly) | [`.buildkite/cuda/test-weekly.yml`](../../../../.buildkite/cuda/test-weekly.yml) | **Invalid parameters Test · H100** / **· L4** — sweeps `tests/dfx/reliability/invalid_param_test/` |
+| L1, L2 | [`.buildkite/test-ready.yml`](../../../../.buildkite/test-ready.yml) | Steps prefixed **Omni ·**, **TTS ·**, **Diffusion ·** under **E2E Test** — **`source_file_dependencies` required** |
+| L3 | [`.buildkite/test-merge.yml`](../../../../.buildkite/test-merge.yml) | Per-model E2E steps; `-m "advanced_model and …"` — **`source_file_dependencies` required** |
+| L4 | [`.buildkite/test-nightly.yml`](../../../../.buildkite/test-nightly.yml) | **Omni / TTS / Diffusion X2I(&A&T) / X2V** — each group may have **Function**, **Accuracy**, **Perf**, **Doc** steps |
+| Invalid param (weekly) | [`.buildkite/test-weekly.yml`](../../../../.buildkite/test-weekly.yml) | **Invalid parameters Test · H100** / **· L4** — sweeps `tests/dfx/reliability/invalid_param_test/` |
 
 ### `source_file_dependencies` (E2E Test only — ready & merge)
 
@@ -380,7 +357,7 @@ Common patterns in `test-ready.yml` / `test-merge.yml`:
 
 When extending an existing step (e.g. add `tests/e2e/offline_inference/test_qwen_image.py` to **Diffusion · Qwen Image Test**), update `source_file_dependencies` and `commands` only; **keep the existing `agents` + `plugins` block unchanged** unless the hardware tier changes.
 
-[`.buildkite/cuda/pipeline.yml`](../../../../.buildkite/cuda/pipeline.yml) decides **which** child file is uploaded. To run **L3** on a feature branch, comment out `if: build.branch == "main"` on the merge upload step. To run **L4** without `NIGHTLY=1`, comment out the nightly upload step’s `if: build.env("NIGHTLY") == "1"` and the same `if` on relevant steps inside `test-nightly.yml`. Revert such edits before merging.
+The root [`.buildkite/pipeline.yml`](../../../../.buildkite/pipeline.yml) decides **which** child file is uploaded. To run **L3** on a feature branch, comment out `if: build.branch == "main"` on the merge upload step. To run **L4** without `NIGHTLY=1`, comment out the nightly upload step’s `if: build.env("NIGHTLY") == "1"` and the same `if` on relevant steps inside `test-nightly.yml`. Revert such edits before merging.
 
 Platform-specific pipelines (e.g. AMD) follow the same level → file pairing under `.buildkite/`.
 

.github/CODEOWNERS

@@ -16,36 +16,36 @@
 /.claude/ @hsliuustc0106
 
 # Tests
-/tests/ @yenuo26 @NickCao
+/tests/ @yenuo26
 
 # CI
-/.buildkite/ @yenuo26 @congw729 @NickCao
+/.buildkite/ @yenuo26 @congw729
 
 # Engine and model_executor
-/vllm_omni/core/ @tzhouam @yinpeiqi @fake0fan @Sy0307
-/vllm_omni/engine/ @tzhouam @yinpeiqi @fake0fan @Sy0307
-/vllm_omni/entrypoints/ @tzhouam @yinpeiqi @fake0fan @alex-jw-brooks @NickCao
+/vllm_omni/core/ @tzhouam @yinpeiqi @fake0fan
+/vllm_omni/engine/ @tzhouam @yinpeiqi @fake0fan
+/vllm_omni/entrypoints/ @tzhouam @yinpeiqi @fake0fan
 /vllm_omni/model_executor/ @tzhouam @gcanlin
 /vllm_omni/worker/ @tzhouam @gcanlin
 
 # Diffusion
-/vllm_omni/diffusion/ @Isotr0py @princepride @SamitHuang @wtomin @ZJY0516 @RuixiangMa @david6666666 @xuechendi
+/vllm_omni/diffusion/ @Isotr0py @princepride @SamitHuang @wtomin @ZJY0516 @RuixiangMa @david6666666
 /vllm_omni/lora/ @princepride @SamitHuang @wtomin @ZJY0516 @RuixiangMa @yuanheng-zhao
 
 # Diffusion CPU offloading
-/vllm_omni/diffusion/offloader/ @Isotr0py @princepride @SamitHuang @wtomin @ZJY0516 @RuixiangMa @yuanheng-zhao @xuechendi
+/vllm_omni/diffusion/offloader/ @Isotr0py @princepride @SamitHuang @wtomin @ZJY0516 @RuixiangMa @yuanheng-zhao
 
 # Omni and TTS support
 /vllm_omni/transformers_utils/ @tzhouam @yuanheng-zhao
-/vllm_omni/model_executor/models/ @linyueqian @ZeldaHuang @princepride @yuanheng-zhao @amy-why-3459 @Bounty-hunter @Sy0307 @gcanlin @NickCao
+/vllm_omni/model_executor/models/ @linyueqian @ZeldaHuang @princepride @yuanheng-zhao @amy-why-3459 @Bounty-hunter
 
 # Benchmarks
 /benchmarks/accuracy/ @david6666666
 /benchmarks/build_dataset/ @linyueqian @yuanheng-zhao
-/benchmarks/diffusion/ @Isotr0py @princepride @SamitHuang @wtomin @ZJY0516 @RuixiangMa @xuechendi
+/benchmarks/diffusion/ @Isotr0py @princepride @SamitHuang @wtomin @ZJY0516 @RuixiangMa
 /benchmarks/distributed/ @princepride
 /benchmarks/glm_image/ @princepride @JaredforReal
-/benchmarks/tts/ @linyueqian @ZeldaHuang @princepride @Sy0307 @gcanlin
+/benchmarks/tts/ @linyueqian @ZeldaHuang @princepride
 
 # Usability and Observability
 /vllm_omni/deploy/ @tzhouam @lishunyang12
@@ -53,13 +53,13 @@
 /vllm_omni/profiler/ @gcanlin
 
 # ComfyUI
-/apps/ @fhfuih @alex-jw-brooks
+/apps/ @fhfuih
 
 # Large Scale Serving
 /vllm_omni/distributed/ @princepride @yuanheng-zhao @xuechendi @natureofnature @fake0fan
 
 # Quantization and configuration
-/vllm_omni/config/ @lishunyang12 @alex-jw-brooks
+/vllm_omni/config/ @lishunyang12
 /vllm_omni/quantization/ @david6666666 @Isotr0py @lishunyang12
 
 # Hardware plugins and accelerators

.github/ISSUE_TEMPLATE/400-bug-report.yml

@@ -19,7 +19,6 @@ body:
 
       Consider redacting or replacing sensitive values with placeholders like `<YOUR_TOKEN_HERE>` when sharing configuration or code examples.
 - type: textarea
-  id: current-environment
   attributes:
     label: Your current environment
     description: |
@@ -43,7 +42,6 @@ body:
     required: true
 
 - type: textarea
-  id: code-version
   attributes:
     label: Your code version
     description: |
@@ -68,7 +66,6 @@ body:
     required: true
 
 - type: textarea
-  id: bug-description
   attributes:
     label: 🐛 Describe the bug
     description: |

.github/workflows/pre-commit.yml

@@ -24,9 +24,3 @@ jobs:
           python-version: "3.12"
           cache: 'pip'
       - uses: pre-commit/action@v3.0.1
-        # Don't run the pytest mark check against everything in the CI
-        # for now since a lot of files are still missing them. This check
-        # is used to help contributors remember to add the marks when they
-        # commit.
-        env:
-          SKIP: check-test-ci-coverage

.gitignore

@@ -208,7 +208,6 @@ cache/
 .cache/
 diffusion_cache/
 kv_cache/
-!vllm_omni/experimental/ar_diffusion/kv_cache/
 
 # Logs
 logs/
@@ -242,13 +241,11 @@ datasets/
 *.json
 !tests/dfx/perf/tests/*.json
 !apps/ComfyUI-vLLM-Omni/example_workflows/*.json
-!tests/assets/soulxsinger/*.json
 *.jsonl
 *.parquet
 
 # Output files
 outputs/
-!vllm_omni/outputs/
 results/
 generated/
 output_*/
@@ -259,6 +256,7 @@ configs/production.yaml
 configs/development.yaml
 
 # Docker
+.dockerignore
 Dockerfile.dev
 discussion
 tmp_test

.pre-commit-config.yaml

@@ -54,13 +54,6 @@ repos:
       verbose: true
       stages: [commit-msg]
 
-    - id: check-test-ci-coverage
-      name: Ensure test files have pytestmarks
-      entry: python tools/pre_commit/check_test_marks.py
-      language: python
-      types: [python]
-      files: ^tests/
-
     # Keep `suggestion` last
     - id: suggestion
       name: Suggestion

README.md

@@ -16,10 +16,13 @@ Easy, fast, and cheap omni-modality model serving for everyone
 ---
 
 *Latest News* 🔥
-- [2026/08] We released [0.26.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.26.0) - aligned with the vLLM 0.26 release line, featuring [MiniMax H3](recipes/MiniMaxAI/MiniMax-H3.md) joint video/audio generation, an experimental full-duplex realtime runtime for [MiniCPM-o 4.5](recipes/OpenBMB/MiniCPM-o-4_5.md), distributed layerwise diffusion offload, and broader model, hardware, streaming, TTS, and quantization support.
-- [2026/07] We released [0.24.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.24.0) - aligned with the vLLM 0.24 release line, expanding production-ready coverage across TTS, speech, diffusion, image/video generation, and robot-policy serving, with major Omni stage runtime refactoring, diffusion request-level batching, async output materialization, quantization/cache/memory improvements, and broad CUDA/ROCm/XPU/NPU support.
-- [2026/06] Starting with [0.14.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.14.0), vLLM-Omni publishes a stable release aligned with every even-numbered upstream vLLM minor version. [0.16.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.16.0), [0.18.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.18.0), [0.20.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.20.0), and [0.22.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.22.0) continued this cadence, expanding omni and world-model support with [NVIDIA Cosmos3](recipes/cosmos3/Cosmos3-Nano.md) and DreamZero, adding models such as MiniCPM-o 4.5, MOSS-TTS, and Lance, and advancing TTS, diffusion, distributed execution, quantization, RL integration through [VeRL-Omni](https://github.com/verl-project/verl-omni), and CUDA/ROCm/MUSA/NPU/XPU coverage.
+- [2026/06] We released [0.22.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.22.0) - an **omnimodal world-model** release aligned with vLLM 0.22, featuring [Nvidia Cosmos3](recipes/cosmos3/Cosmos3-Nano.md)/DreamZero world model support, expanded quantization coverage across Blackwell/NPU/XPU, TTS production improvements, new models including MiniCPM-o 4.5, MOSS-TTS, and Lance, plus RL integration with [VeRL-Omni](https://github.com/verl-project/verl-omni).
+- [2026/05] We released [0.20.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.20.0) - refreshes the serving/runtime stack for large-scale omni workloads, and improves diffusion model performance, quantization, and hardware readiness across CUDA, ROCm, MUSA, NPU, and XPU backends.
+- [2026/03] We released [0.18.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.18.0) - strengthens the core runtime through a large entrypoint refactor and scheduler/runtime cleanups, expands unified quantization and diffusion execution, broadens multimodal model coverage, and improves production readiness across audio, omni, image, video, RL, and multi-platform deployments.
 - [2026/03] Check out our first public [project deepdive](https://youtu.be/sgwNfsNnR9I) at the vLLM Hong Kong Meetup!
+- [2026/03] **[vllm-omni-skills](https://github.com/hsliuustc0106/vllm-omni-skills)** is a community-driven collection of AI assistant skills that help developers work with vLLM-Omni more effectively. These skills can be used with popular agentic AI coding assistants like **Cursor IDE**, **Claude**, **Codex**, and more.
+- [2026/02] We released [0.16.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.16.0) - A major alignment + capability release that rebases onto **upstream vLLM v0.16.0** and significantly expands performance, distributed execution, and production readiness across **Qwen3-Omni / Qwen3-TTS**, **Bagel**, **MiMo-Audio**, **GLM-Image** and the **Diffusion (DiT) image/video stack**—while also improving platform coverage (CUDA / ROCm / NPU / XPU), CI quality, and documentation.
+- [2026/02] We released [0.14.0](https://github.com/vllm-project/vllm-omni/releases/tag/v0.14.0) - This is the first **stable release** of vLLM-Omni that expands Omni’s diffusion / image-video generation and audio / TTS stack, improves distributed execution and memory efficiency, and broadens platform/backend coverage (GPU/ROCm/NPU/XPU). It also brings meaningful upgrades to serving APIs, profiling & benchmarking, and overall stability. Please check our latest [paper](https://arxiv.org/abs/2602.02204) for architecture design and performance results.
 - [2025/11] vLLM community officially released [vllm-project/vllm-omni](https://github.com/vllm-project/vllm-omni) in order to support omni-modality models serving.
 
 ---
@@ -28,9 +31,9 @@ Easy, fast, and cheap omni-modality model serving for everyone
 
 [vLLM](https://github.com/vllm-project/vllm) was originally designed to support large language models for text-based autoregressive generation tasks. vLLM-Omni is a framework that extends its support for omni-modality model inference and serving:
 
-- **Omni-modality**: Text, image, audio, video, and action data processing
+- **Omni-modality**: Text, image, video, and audio data processing
 - **Non-autoregressive Architectures**: extend the AR support of vLLM to Diffusion Transformers (DiT) and other parallel generation models
-- **Heterogeneous outputs**: from traditional text generation to multimodal and action outputs
+- **Heterogeneous outputs**: from traditional text generation to multimodal outputs
 
 <p align="center">
   <picture>
@@ -51,14 +54,12 @@ vLLM-Omni is flexible and easy to use with:
 - Tensor, pipeline, data and expert parallelism support for distributed inference
 - Streaming outputs
 - OpenAI-compatible API server
-- Full-duplex realtime serving with streaming audio input and output (experimental)
 
 vLLM-Omni seamlessly supports most popular open-source models on HuggingFace, including:
 
-- **Omni-modality models** (e.g. Qwen3-Omni, MiniCPM-o 4.5, Cosmos3, HunyuanImage, BAGEL)
+- **Omni-modality models** (e.g. Qwen3-Omni, Cosmos, HunyuanImage, BAGEL)
 - **TTS models** (e.g. Qwen3-TTS, VoxCPM2, Ming-Omni-TTS, CosyVoice3)
-- **Diffusion models** — image, video, and audio generation (e.g. MiniMax H3, Qwen-Image, Wan2.2, FLUX)
-- **Robot-policy and action models** (e.g. GR00T-N1.7, DreamZero-DROID, InternVLA-A1, Cosmos3 action policy)
+- **Diffusion models** — image, video, and audio generation (e.g. Qwen-Image, Wan2.2, FLUX)
 
 ## Getting Started
 

apps/ComfyUI-vLLM-Omni/README.md

@@ -47,7 +47,7 @@ If no, check your shell running the ComfyUI process. There may be some error mes
 This extension offers the following nodes based on the output modalities (at **ComfyUI sidebar -> Node Library**):
 
 - **Generate Image** for text-to-image and image-to-image tasks
-- **Generate Video** for text-to-video, first-frame/image-to-video, and reference-conditioned video
+- **Generate Video** for text-to-video and image-to-video tasks
 - **Multimodality Understanding** for multimodality-to-text and multimodality-to-audio tasks
 - **TTS** and **TTS Voice Clone** for TTS tasks
 
@@ -105,7 +105,7 @@ You can configure per-stage sampling parameters for multi-stage models.
 > [!TIP]
 > The node automatically choose text-to-image or image-to-image API endpoints depending on whether you connect an image input or not.
 
-### Text-to-video and image-to-video generation (e.g., Wan, MiniMax-H3)
+### Text-to-video and image-to-video generation (e.g., Wan)
 
 (Also available at **ComfyUI sidebar->Template->vLLM-Omni->vLLM-Omni Video Generation**)
 
@@ -117,11 +117,7 @@ You can configure per-stage sampling parameters for multi-stage models.
 </p>
 
 > [!TIP]
-> Connect a **frame** image for first-frame / image-to-video (e.g., Wan I2V, MiniMax-H3 FL2VA).
->
-> For reference-conditioned generation (MiniMax-H3 Ref2VA), connect a **Video References** node instead.
->
-> Do not use `frame` and `references` together. Task routing is automatic from which inputs you connect.
+> The node automatically choose text-to-video or image-to-video API endpoints depending on whether you connect an image input or not.
 
 ### TTS (e.g., Qwen TTS series)
 

apps/ComfyUI-vLLM-Omni/__init__.py

@@ -15,13 +15,11 @@ from .comfyui_vllm_omni.nodes import (
     VLLMOmniDiffusionSampling,
     VLLMOmniGenerateImage,
     VLLMOmniGenerateVideo,
-    VLLMOmniMiniMaxH3Params,
     VLLMOmniQwenTTSParams,
     VLLMOmniRemoteLoRA,
     VLLMOmniSamplingParamsList,
     VLLMOmniTTS,
     VLLMOmniUnderstanding,
-    VLLMOmniVideoReferences,
     VLLMOmniVoiceClone,
     VLLMOmniWanParams,
 )
@@ -34,7 +32,6 @@ NODE_CLASS_MAPPINGS = {
     "VLLMOmniUnderstanding": VLLMOmniUnderstanding,
     "VLLMOmniTTS": VLLMOmniTTS,
     "VLLMOmniVoiceClone": VLLMOmniVoiceClone,
-    "VLLMOmniVideoReferences": VLLMOmniVideoReferences,
     # === Params ===
     "VLLMOmniARSampling": VLLMOmniARSampling,
     "VLLMOmniDiffusionSampling": VLLMOmniDiffusionSampling,
@@ -42,7 +39,6 @@ NODE_CLASS_MAPPINGS = {
     "VLLMOmniRemoteLoRA": VLLMOmniRemoteLoRA,
     "VLLMOmniQwenTTSParams": VLLMOmniQwenTTSParams,
     "VLLMOmniWanParams": VLLMOmniWanParams,
-    "VLLMOmniMiniMaxH3Params": VLLMOmniMiniMaxH3Params,
 }
 
 # A dictionary that contains the friendly/humanly readable titles for the nodes
@@ -53,7 +49,6 @@ NODE_DISPLAY_NAME_MAPPINGS = {
     "VLLMOmniUnderstanding": "Multimodality Understanding",
     "VLLMOmniTTS": "TTS (Text to Speech)",
     "VLLMOmniVoiceClone": "TTS Voice Cloning",
-    "VLLMOmniVideoReferences": "Video References",
     # === Params ===
     "VLLMOmniARSampling": "AR Sampling Params",
     "VLLMOmniDiffusionSampling": "Diffusion Sampling Params",
@@ -61,7 +56,6 @@ NODE_DISPLAY_NAME_MAPPINGS = {
     "VLLMOmniRemoteLoRA": "LoRA",
     "VLLMOmniQwenTTSParams": "Qwen TTS Params",
     "VLLMOmniWanParams": "Wan Video Params",
-    "VLLMOmniMiniMaxH3Params": "MiniMax-H3 Video Params",
 }
 
 WEB_DIRECTORY = "./web"

apps/ComfyUI-vLLM-Omni/comfyui_vllm_omni/nodes.py

@@ -10,9 +10,7 @@ from .utils.types import (
     AudioFormat,
     AutoregressionSamplingParams,
     DiffusionSamplingParams,
-    MiniMaxH3ModelSpecificParams,
     QwenTTSModelSpecificParams,
-    VideoReferences,
     WanModelSpecificParams,
 )
 from .utils.validators import (
@@ -165,8 +163,7 @@ class VLLMOmniGenerateVideo(_VLLMOmniGenerateBase):
                 "num_frames": ("INT", {"default": 41, "min": 1}),
             },
             "optional": {
-                "frame": ("IMAGE",),
-                "references": ("VIDEO_REFERENCES",),
+                "image": ("IMAGE",),
                 "sampling_params": ("SAMPLING_PARAMS",),
                 "lora": ("REMOTE_LORA",),
                 "model_params": ("VIDEO_PARAMS",),
@@ -177,15 +174,6 @@ class VLLMOmniGenerateVideo(_VLLMOmniGenerateBase):
     RETURN_NAMES = ("video",)
     FUNCTION = "generate"
 
-    @classmethod
-    def VALIDATE_INPUTS(cls, url, model, frame=None, references=None, **_kwargs) -> str | Literal[True]:
-        base = super().VALIDATE_INPUTS(url, model)
-        if base is not True:
-            return base
-        if frame is not None and references is not None:
-            return "Provide only one of frame or references, not both."
-        return True
-
     async def generate(
         self,
         url: str,
@@ -196,8 +184,7 @@ class VLLMOmniGenerateVideo(_VLLMOmniGenerateBase):
         fps: int,
         num_frames: int,
         negative_prompt: str | None = None,
-        frame: torch.Tensor | None = None,
-        references: dict | None = None,
+        image: torch.Tensor | None = None,
         sampling_params: dict | list[dict] | None = None,
         model_params: dict | None = None,
         lora: dict | None = None,
@@ -220,16 +207,14 @@ class VLLMOmniGenerateVideo(_VLLMOmniGenerateBase):
 
         if sampling_params is not None:
             sampling_params.pop("type", None)  # internal fields
-        # model_params["type"] is also an internal field, but is used in api_client
-        # if model_params is not None:
-        #     model_params.pop("type", None)  # internal fields
+        if model_params is not None:
+            model_params.pop("type", None)  # internal fields
 
         client = VLLMOmniClient(url)
         output = await client.generate_video(
             model=model,
             prompt=prompt,
-            frame=frame,  # frame present => fl2va / Wan I2V
-            references=references,
+            image=image,  # image present => i2v, absent => t2v
             width=width,
             height=height,
             num_frames=num_frames,
@@ -749,78 +734,3 @@ class VLLMOmniWanParams:
 
     def get_params(self, **kwargs):
         return (WanModelSpecificParams(kwargs),)
-
-
-class VLLMOmniMiniMaxH3Params:
-    @classmethod
-    def INPUT_TYPES(cls):
-        return {
-            "required": {
-                "audio_flow_shift": (
-                    "FLOAT",
-                    {"default": 3.0, "min": 0.0, "max": 100.0, "step": 0.1},
-                ),
-                "flow_shift": (
-                    "FLOAT",
-                    {"default": 12.0, "min": 0.0, "max": 100.0, "step": 0.1},
-                ),
-            }
-        }
-
-    RETURN_TYPES = ("VIDEO_PARAMS",)
-    RETURN_NAMES = ("MiniMax-H3 video params",)
-    FUNCTION = "get_params"
-    CATEGORY = "vLLM-Omni/Video Params"
-
-    def get_params(self, **kwargs):
-        params = MiniMaxH3ModelSpecificParams(kwargs)
-        params["type"] = "minimax_h3"
-        return (params,)
-
-
-class VLLMOmniVideoReferences:
-    @classmethod
-    def INPUT_TYPES(cls):
-        return {
-            "required": {},
-            "optional": {
-                "image_1": ("IMAGE",),
-                "image_2": ("IMAGE",),
-                "audio_1": ("AUDIO",),
-                "audio_2": ("AUDIO",),
-                "video_1": ("VIDEO",),
-                "video_2": ("VIDEO",),
-            },
-        }
-
-    RETURN_TYPES = ("VIDEO_REFERENCES",)
-    RETURN_NAMES = ("references",)
-    FUNCTION = "get_references"
-    CATEGORY = "vLLM-Omni"
-
-    def get_references(
-        self,
-        image_1: torch.Tensor | None = None,
-        image_2: torch.Tensor | None = None,
-        audio_1: AudioInput | None = None,
-        audio_2: AudioInput | None = None,
-        video_1: VideoInput | None = None,
-        video_2: VideoInput | None = None,
-        **kwargs,
-    ):
-        if kwargs:
-            logger.info("Uncaught kwargs: %s", kwargs)
-        refs = VideoReferences()
-        if image_1 is not None:
-            refs["image_1"] = image_1
-        if image_2 is not None:
-            refs["image_2"] = image_2
-        if audio_1 is not None:
-            refs["audio_1"] = audio_1
-        if audio_2 is not None:
-            refs["audio_2"] = audio_2
-        if video_1 is not None:
-            refs["video_1"] = video_1
-        if video_2 is not None:
-            refs["video_2"] = video_2
-        return (refs,)

apps/ComfyUI-vLLM-Omni/comfyui_vllm_omni/utils/api_client.py

@@ -24,7 +24,6 @@ from .format import (
     image_tensor_to_base64,
     image_tensor_to_png_bytes,
     video_to_base64,
-    video_to_bytes,
 )
 from .logger import get_logger, pretty_printer
 from .models import lookup_model_spec
@@ -64,11 +63,7 @@ async def url_bytes(session: aiohttp.ClientSession, url: str, verb: str = "get",
 
 class VLLMOmniClient:
     def __init__(
-        self,
-        base_url: str,
-        timeout: float | None = None,
-        poll_interval: float = 5.0,
-        max_poll_duration: float = 60 * 30,
+        self, base_url: str, timeout: float | None = None, poll_interval: float = 5.0, max_poll_duration: float = 60 * 5
     ):
         self.base_url = base_url
         self.timeout = aiohttp.ClientTimeout(total=timeout)
@@ -266,17 +261,12 @@ class VLLMOmniClient:
         num_frames: int,
         fps: int,
         negative_prompt: str | None = None,
-        frame: torch.Tensor | None = None,
-        references: dict | None = None,
+        image: torch.Tensor | None = None,
         sampling_params: dict | None = None,
         model_params: dict | None = None,
         lora: dict | None = None,
-        **extra_params,
+        **extra_body,
     ) -> VideoInput:
-        if frame is not None and references is not None:
-            raise ValueError("Provide only one of frame or references, not both.")
-
-        # === regular payload fields ===
         form = aiohttp.FormData()
         form.add_field("model", model)
         form.add_field("prompt", prompt)
@@ -289,91 +279,23 @@ class VLLMOmniClient:
         if sampling_params is not None:
             for k, v in sampling_params.items():
                 form.add_field(k, str(v))
+        if model_params is not None:
+            for k, v in model_params.items():
+                form.add_field(k, str(v))
         if lora is not None:
             form.add_field("lora", json.dumps(lora, ensure_ascii=False))
+        if extra_body:
+            form.add_field("extra_body", json.dumps(extra_body, ensure_ascii=False))
 
-        # === multimodal inputs (first-last-frames, references, etc.) ===
-        input_reference_image: torch.Tensor | None = None
-        audio_reference: AudioInput | None = None
-        reference_videos: list[VideoInput] = []
-        video_task: str | None = None
-
-        if frame is not None:
-            input_reference_image = frame
-            video_task = "fl2va"
-        elif references is not None:
-            images = [references[k] for k in ("image_1", "image_2") if k in references and references[k] is not None]
-            audios = [references[k] for k in ("audio_1", "audio_2") if k in references and references[k] is not None]
-            videos = [references[k] for k in ("video_1", "video_2") if k in references and references[k] is not None]
-
-            if not images and not audios and not videos:
-                raise ValueError("references is empty; connect at least one image/audio/video.")
-
-            if videos:
-                if images or audios:
-                    raise ValueError(
-                        "Video references cannot be combined with image or audio references. "
-                        "Connect only video_1/video_2 for multi-video Ref2VA."
-                    )
-                reference_videos = videos
-                video_task = "ref2va"
-            elif len(images) == 1 and len(audios) == 1:
-                input_reference_image = images[0]
-                audio_reference = audios[0]
-                video_task = "ref2va"
-            else:
-                raise ValueError(
-                    "Invalid references combination. Supported modes: "
-                    "(1) one or more videos only, or (2) exactly one image and one audio."
-                )
-        else:
-            video_task = "t2va"
-
-        if input_reference_image is not None:
+        if image is not None:
             image_filename = "image.png"  # Required for multipart form
             form.add_field(
                 "input_reference",
-                image_tensor_to_png_bytes(input_reference_image, image_filename),
+                image_tensor_to_png_bytes(image, image_filename),
                 filename=image_filename,
                 content_type="image/png",
             )
 
-        if audio_reference is not None:
-            form.add_field(
-                "audio_reference",
-                json.dumps({"audio_url": audio_to_base64(audio_reference)}, ensure_ascii=False),
-            )
-
-        for idx, video in enumerate(reference_videos, start=1):
-            video_filename = f"reference_{idx}.mp4"
-            form.add_field(
-                "input_references",
-                video_to_bytes(video, video_filename),
-                filename=video_filename,
-                content_type="video/mp4",
-            )
-
-        # === model specific params. Either use a specialized builder, or add flattened fields as-is ===
-        if model_params is not None:
-            model_params = dict(model_params)
-            model_params.pop("type", None)
-
-        spec, _ = lookup_model_spec(model)
-        params_builder = spec.get("params_builder") if spec else None
-        if params_builder is not None:
-            form_fields = params_builder(
-                model_params or {},
-                extra_params={**extra_params, "task": video_task},
-            )
-            for k, v in form_fields.items():
-                form.add_field(k, v if isinstance(v, str) else str(v))
-        else:
-            if model_params is not None:
-                for k, v in model_params.items():
-                    form.add_field(k, str(v))
-            if extra_params:
-                form.add_field("extra_params", json.dumps(extra_params, ensure_ascii=False))
-
         async with aiohttp.ClientSession(timeout=self.timeout) as session:
             # Start the video generation job
             url = f"{self.base_url}/videos"

apps/ComfyUI-vLLM-Omni/comfyui_vllm_omni/utils/models.py

@@ -1,12 +1,7 @@
-import json
 import re
-from typing import Any
 
-from .logger import get_logger
 from .types import Modality, ModelMode, Spec
 
-logger = get_logger(__name__)
-
 
 def _bagel_payload_preprocessor(payload: dict) -> dict:
     try:
@@ -34,30 +29,6 @@ def _qwen25_payload_preprocessor(payload: dict) -> dict:
     return payload
 
 
-def _minimaxh3_params_builder(
-    model_params: dict[str, Any],
-    *,
-    extra_params: dict[str, Any],
-) -> dict[str, Any]:
-    """Build multipart form fields for MiniMax-H3 from model_params + routed task."""
-    params = dict(model_params)
-    params.pop("type", None)
-    form_fields: dict[str, Any] = {}
-    merged_extra_params: dict[str, Any] = dict(extra_params or {})
-
-    if "flow_shift" in params:
-        form_fields["flow_shift"] = params.pop("flow_shift")
-    for key in ("audio_flow_shift",):
-        if key in params:
-            merged_extra_params[key] = params.pop(key)
-    if params:
-        logger.warning("Unused MiniMax-H3 model params ignored: %s", sorted(params))
-
-    if merged_extra_params:
-        form_fields["extra_params"] = json.dumps(merged_extra_params, ensure_ascii=False)
-    return form_fields
-
-
 _MODEL_PIPELINE_SPECS: dict[str, Spec] = {
     r"BAGEL-7B-MoT": {
         "stages": [
@@ -100,21 +71,6 @@ _MODEL_PIPELINE_SPECS: dict[str, Spec] = {
             }
         ],
     },
-    r"MiniMax-H3(/FL2VA|/Ref2VA)?": {
-        "stages": ["diffusion"],
-        "modes": [
-            {
-                "mode": ModelMode.VIDEO_GENERATION,
-                "input_modalities": [
-                    Modality.TEXT,
-                    Modality.IMAGE,
-                    Modality.VIDEO,
-                    Modality.AUDIO,
-                ],
-            }
-        ],
-        "params_builder": _minimaxh3_params_builder,
-    },
 }
 # Convert dict keys to regex patterns
 MODEL_PIPELINE_SPECS: dict[re.Pattern, Spec] = {}
@@ -124,10 +80,12 @@ del _MODEL_PIPELINE_SPECS
 
 
 def lookup_model_spec(model: str) -> tuple[Spec | None, str | None]:
-    normalized = model.rstrip("/").lstrip("/")
-    lookup_key = normalized.split("/", 1)[-1]
+    try:
+        last_component = model.rstrip("/").rsplit("/", 1)[-1]
+    except IndexError:
+        last_component = model
     for pattern, spec in MODEL_PIPELINE_SPECS.items():
-        if pattern.search(lookup_key):
+        if pattern.search(last_component):
             return spec, pattern.pattern
     return None, None
 
@@ -143,9 +101,6 @@ if __name__ == "__main__":
         "Qwen/Qwen3-Omni-30B-A3B-Instruct",
         "Custom/Path/UnknownModel-Instruct",
         "Not/Matching/Anything",
-        "/path/to/MiniMax-H3/FL2VA",
-        "/path/to/MiniMax-H3/Ref2VA",
-        "MiniMaxAI/MiniMax-H3",
     ]
 
     test_payload = {"messages": [{"role": "user", "content": "prompt"}]}
@@ -157,13 +112,7 @@ if __name__ == "__main__":
             if preprocessor := spec.get("payload_preprocessor"):
                 result = preprocessor(test_payload)
                 print(f"✓ {path:<40} → {result}")
-            elif params_builder := spec.get("params_builder"):
-                result = params_builder(
-                    {"audio_flow_shift": 3.0, "flow_shift": 12.0, "type": "minimax_h3"},
-                    extra_params={"task": "t2va"},
-                )
-                print(f"✓ {path:<40} → {result}")
             else:
-                print(f"✓ {path:<40} → No preprocessor/params_builder")
+                print(f"✓ {path:<40} → No preprocessor")
         else:
             print(f"✗ {path:<40} → No match")

apps/ComfyUI-vLLM-Omni/comfyui_vllm_omni/utils/types.py

@@ -1,8 +1,12 @@
 from collections.abc import Callable
 from enum import Enum, auto
-from typing import Any, Literal, TypeAlias
-
-from typing_extensions import NotRequired, TypedDict
+from typing import (
+    Any,
+    Literal,
+    NotRequired,
+    TypeAlias,
+    TypedDict,
+)
 
 AudioFormat: TypeAlias = Literal["mp3", "opus", "aac", "flac", "wav", "pcm"]
 
@@ -23,14 +27,6 @@ class WanModelSpecificParams(dict):
     pass
 
 
-class MiniMaxH3ModelSpecificParams(dict):
-    pass
-
-
-class VideoReferences(dict):
-    pass
-
-
 class ModelMode(Enum):
     IMAGE_GENERATION = auto()
     VIDEO_GENERATION = auto()
@@ -51,11 +47,9 @@ class ModelModeSpec(TypedDict):
 
 
 PayloadPreprocessor: TypeAlias = Callable[[dict[str, Any]], dict[str, Any]]
-ParamsBuilder: TypeAlias = Callable[..., dict[str, Any]]
 
 
 class Spec(TypedDict):
     stages: list[Literal["diffusion", "autoregression"]]
     modes: list[ModelModeSpec]
     payload_preprocessor: NotRequired[PayloadPreprocessor]
-    params_builder: NotRequired[ParamsBuilder]

apps/ComfyUI-vLLM-Omni/example_workflows/vLLM-Omni Video Generation.json

@@ -1 +1,513 @@
-{"id":"3aea1d08-7de5-4997-ba71-b2f66ad72a21","revision":0,"last_node_id":24,"last_link_id":35,"nodes":[{"id":13,"type":"SaveVideo","pos":[1315.729153459605,-137.92038948394355],"size":[374,313.7692307692308],"flags":{},"order":9,"mode":0,"inputs":[{"localized_name":"video","name":"video","type":"VIDEO","link":32},{"localized_name":"filename_prefix","name":"filename_prefix","type":"STRING","widget":{"name":"filename_prefix"},"link":null},{"localized_name":"format","name":"format","type":"COMBO","widget":{"name":"format"},"link":null},{"localized_name":"codec","name":"codec","type":"COMFY_DYNAMICCOMBO_V3","widget":{"name":"codec"},"link":null}],"outputs":[{"localized_name":"video","name":"video","type":"VIDEO","links":null}],"properties":{},"widgets_values":["video/ComfyUI","auto","auto"]},{"id":10,"type":"MarkdownNote","pos":[183.74636797766794,-571.2529158046766],"size":[240.3326478922474,136.3505820791881],"flags":{},"order":0,"mode":0,"inputs":[],"outputs":[],"title":"Note: Task and Input","properties":{},"widgets_values":["vLLM-Omni nodes are categorized based on the output modality. The \"Generate Video\" node supports both *text-to-video*, *frame-to-video*, and *reference-to-video* generation. The node will create corresponding payload depending on whether an input image is present or not."],"color":"#432","bgcolor":"#000"},{"id":4,"type":"LoadImage","pos":[452.07189605801125,-569.7242536425791],"size":[270,314],"flags":{},"order":1,"mode":0,"inputs":[{"localized_name":"image","name":"image","type":"COMBO","widget":{"name":"image"},"link":null},{"localized_name":"choose file to upload","name":"upload","type":"IMAGEUPLOAD","widget":{"name":"upload"},"link":null}],"outputs":[{"localized_name":"IMAGE","name":"IMAGE","type":"IMAGE","links":[29]},{"localized_name":"MASK","name":"MASK","type":"MASK","links":null}],"properties":{"Node name for S&R":"LoadImage"},"widgets_values":["rap.png","image"]},{"id":23,"type":"LoadImage","pos":[9.206110246057657,-7.6955291509170785],"size":[270,314],"flags":{},"order":2,"mode":0,"inputs":[{"localized_name":"image","name":"image","type":"COMBO","widget":{"name":"image"},"link":null},{"localized_name":"choose file to upload","name":"upload","type":"IMAGEUPLOAD","widget":{"name":"upload"},"link":null}],"outputs":[{"localized_name":"IMAGE","name":"IMAGE","type":"IMAGE","links":[34]},{"localized_name":"MASK","name":"MASK","type":"MASK","links":null}],"properties":{"Node name for S&R":"LoadImage"},"widgets_values":["white-cat.jpg","image"]},{"id":22,"type":"MarkdownNote","pos":[-9.783113411416666,-148.8790300038605],"size":[297.3012007349086,88],"flags":{},"order":3,"mode":0,"inputs":[],"outputs":[],"title":"Note: Reference to video (r2v)","properties":{},"widgets_values":["For r2v tasks, connect this Video Reference block to the main Generate Video block. Then disconnect the \"frame\" input above."],"color":"#432","bgcolor":"#000"},{"id":24,"type":"LoadAudio","pos":[317.9929742385805,-141.44821500065095],"size":[282.798828125,136],"flags":{},"order":4,"mode":0,"inputs":[{"localized_name":"audio","name":"audio","type":"COMBO","widget":{"name":"audio"},"link":null},{"localized_name":"audioUI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null},{"localized_name":"choose file to upload","name":"upload","type":"AUDIOUPLOAD","widget":{"name":"upload"},"link":null}],"outputs":[{"localized_name":"AUDIO","name":"AUDIO","type":"AUDIO","links":[35]}],"properties":{"Node name for S&R":"LoadAudio"},"widgets_values":["nothing-beats-a-jet2-holiday.mp3",null,""]},{"id":21,"type":"VLLMOmniVideoReferences","pos":[376.6129172897595,105.41557277556292],"size":[156.9380859375,126],"flags":{},"order":7,"mode":0,"inputs":[{"localized_name":"image_1","name":"image_1","shape":7,"type":"IMAGE","link":34},{"localized_name":"image_2","name":"image_2","shape":7,"type":"IMAGE","link":null},{"localized_name":"audio_1","name":"audio_1","shape":7,"type":"AUDIO","link":35},{"localized_name":"audio_2","name":"audio_2","shape":7,"type":"AUDIO","link":null},{"localized_name":"video_1","name":"video_1","shape":7,"type":"VIDEO","link":null},{"localized_name":"video_2","name":"video_2","shape":7,"type":"VIDEO","link":null}],"outputs":[{"localized_name":"references","name":"references","type":"VIDEO_REFERENCES","links":null}],"properties":{"Node name for S&R":"VLLMOmniVideoReferences"},"widgets_values":[]},{"id":8,"type":"VLLMOmniDiffusionSampling","pos":[83.24620208113915,422.08543477542605],"size":[284.205078125,226],"flags":{},"order":5,"mode":0,"inputs":[{"localized_name":"n","name":"n","type":"INT","widget":{"name":"n"},"link":null},{"localized_name":"num_inference_steps","name":"num_inference_steps","type":"INT","widget":{"name":"num_inference_steps"},"link":null},{"localized_name":"guidance_scale","name":"guidance_scale","type":"FLOAT","widget":{"name":"guidance_scale"},"link":null},{"localized_name":"true_cfg_scale","name":"true_cfg_scale","type":"FLOAT","widget":{"name":"true_cfg_scLine truncated
+{
+    "id": "3aea1d08-7de5-4997-ba71-b2f66ad72a21",
+    "revision": 0,
+    "last_node_id": 14,
+    "last_link_id": 28,
+    "nodes": [
+        {
+            "id": 10,
+            "type": "MarkdownNote",
+            "pos": [
+                227.99306819575278,
+                -231.24143306069843
+            ],
+            "size": [
+                240.3326478922474,
+                136.3505820791881
+            ],
+            "flags": {},
+            "order": 0,
+            "mode": 0,
+            "inputs": [],
+            "outputs": [],
+            "title": "Note: Task and Input",
+            "properties": {},
+            "widgets_values": [
+                "vLLM-Omni nodes are categorized based on the output modality. The \"Generate Video\" node supports both text-to-video generation and image-to-video generation. The node will create corresponding payload depending on whether an input image is present or not."
+            ],
+            "color": "#432",
+            "bgcolor": "#000"
+        },
+        {
+            "id": 4,
+            "type": "LoadImage",
+            "pos": [
+                496.31859627609606,
+                -229.71277089860084
+            ],
+            "size": [
+                270,
+                314
+            ],
+            "flags": {},
+            "order": 1,
+            "mode": 0,
+            "inputs": [
+                {
+                    "localized_name": "image",
+                    "name": "image",
+                    "type": "COMBO",
+                    "widget": {
+                        "name": "image"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "choose file to upload",
+                    "name": "upload",
+                    "type": "IMAGEUPLOAD",
+                    "widget": {
+                        "name": "upload"
+                    },
+                    "link": null
+                }
+            ],
+            "outputs": [
+                {
+                    "localized_name": "IMAGE",
+                    "name": "IMAGE",
+                    "type": "IMAGE",
+                    "links": [
+                        25
+                    ]
+                },
+                {
+                    "localized_name": "MASK",
+                    "name": "MASK",
+                    "type": "MASK",
+                    "links": null
+                }
+            ],
+            "properties": {
+                "Node name for S&R": "LoadImage"
+            },
+            "widgets_values": [
+                "cute-cat.jpg",
+                "image"
+            ]
+        },
+        {
+            "id": 8,
+            "type": "VLLMOmniDiffusionSampling",
+            "pos": [
+                465.5140218220927,
+                148.80072646828987
+            ],
+            "size": [
+                284.205078125,
+                226
+            ],
+            "flags": {},
+            "order": 2,
+            "mode": 0,
+            "inputs": [
+                {
+                    "localized_name": "n",
+                    "name": "n",
+                    "type": "INT",
+                    "widget": {
+                        "name": "n"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "num_inference_steps",
+                    "name": "num_inference_steps",
+                    "type": "INT",
+                    "widget": {
+                        "name": "num_inference_steps"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "guidance_scale",
+                    "name": "guidance_scale",
+                    "type": "FLOAT",
+                    "widget": {
+                        "name": "guidance_scale"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "true_cfg_scale",
+                    "name": "true_cfg_scale",
+                    "type": "FLOAT",
+                    "widget": {
+                        "name": "true_cfg_scale"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "vae_use_slicing",
+                    "name": "vae_use_slicing",
+                    "type": "BOOLEAN",
+                    "widget": {
+                        "name": "vae_use_slicing"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "vae_use_tiling",
+                    "name": "vae_use_tiling",
+                    "type": "BOOLEAN",
+                    "widget": {
+                        "name": "vae_use_tiling"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "seed",
+                    "name": "seed",
+                    "type": "INT",
+                    "widget": {
+                        "name": "seed"
+                    },
+                    "link": null
+                }
+            ],
+            "outputs": [
+                {
+                    "localized_name": "diffusion sampling params",
+                    "name": "diffusion sampling params",
+                    "type": "SAMPLING_PARAMS",
+                    "links": [
+                        27
+                    ]
+                }
+            ],
+            "properties": {
+                "Node name for S&R": "VLLMOmniDiffusionSampling"
+            },
+            "widgets_values": [
+                1,
+                50,
+                1,
+                1,
+                false,
+                false,
+                1992,
+                "randomize"
+            ]
+        },
+        {
+            "id": 14,
+            "type": "VLLMOmniWanParams",
+            "pos": [
+                518.6005521845747,
+                428.4455663719664
+            ],
+            "size": [
+                270,
+                106
+            ],
+            "flags": {},
+            "order": 3,
+            "mode": 0,
+            "inputs": [
+                {
+                    "localized_name": "guidance_scale_2",
+                    "name": "guidance_scale_2",
+                    "type": "FLOAT",
+                    "widget": {
+                        "name": "guidance_scale_2"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "boundary_ratio",
+                    "name": "boundary_ratio",
+                    "type": "FLOAT",
+                    "widget": {
+                        "name": "boundary_ratio"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "flow_shift",
+                    "name": "flow_shift",
+                    "type": "FLOAT",
+                    "widget": {
+                        "name": "flow_shift"
+                    },
+                    "link": null
+                }
+            ],
+            "outputs": [
+                {
+                    "localized_name": "Wan video params",
+                    "name": "Wan video params",
+                    "type": "VIDEO_PARAMS",
+                    "links": [
+                        28
+                    ]
+                }
+            ],
+            "properties": {
+                "Node name for S&R": "VLLMOmniWanParams"
+            },
+            "widgets_values": [
+                4,
+                0.875,
+                5
+            ]
+        },
+        {
+            "id": 12,
+            "type": "VLLMOmniGenerateVideo",
+            "pos": [
+                827.2566336087867,
+                -73.77449831827545
+            ],
+            "size": [
+                400,
+                346
+            ],
+            "flags": {},
+            "order": 4,
+            "mode": 0,
+            "inputs": [
+                {
+                    "localized_name": "image",
+                    "name": "image",
+                    "shape": 7,
+                    "type": "IMAGE",
+                    "link": 25
+                },
+                {
+                    "localized_name": "sampling_params",
+                    "name": "sampling_params",
+                    "shape": 7,
+                    "type": "SAMPLING_PARAMS",
+                    "link": 27
+                },
+                {
+                    "localized_name": "lora",
+                    "name": "lora",
+                    "shape": 7,
+                    "type": "REMOTE_LORA",
+                    "link": null
+                },
+                {
+                    "localized_name": "model_params",
+                    "name": "model_params",
+                    "shape": 7,
+                    "type": "VIDEO_PARAMS",
+                    "link": 28
+                },
+                {
+                    "localized_name": "url",
+                    "name": "url",
+                    "type": "STRING",
+                    "widget": {
+                        "name": "url"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "model",
+                    "name": "model",
+                    "type": "STRING",
+                    "widget": {
+                        "name": "model"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "prompt",
+                    "name": "prompt",
+                    "type": "STRING",
+                    "widget": {
+                        "name": "prompt"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "negative_prompt",
+                    "name": "negative_prompt",
+                    "type": "STRING",
+                    "widget": {
+                        "name": "negative_prompt"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "width",
+                    "name": "width",
+                    "type": "INT",
+                    "widget": {
+                        "name": "width"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "height",
+                    "name": "height",
+                    "type": "INT",
+                    "widget": {
+                        "name": "height"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "fps",
+                    "name": "fps",
+                    "type": "INT",
+                    "widget": {
+                        "name": "fps"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "num_frames",
+                    "name": "num_frames",
+                    "type": "INT",
+                    "widget": {
+                        "name": "num_frames"
+                    },
+                    "link": null
+                }
+            ],
+            "outputs": [
+                {
+                    "localized_name": "video",
+                    "name": "video",
+                    "type": "VIDEO",
+                    "links": [
+                        26
+                    ]
+                }
+            ],
+            "properties": {
+                "Node name for S&R": "VLLMOmniGenerateVideo"
+            },
+            "widgets_values": [
+                "http://localhost:8000/v1",
+                "Wan-AI/Wan2.2-TI2V-5B-Diffusers",
+                "Make the cat in the attached video yawning. Eye closed and mouth wide open. Keep the original realistic style.",
+                "Cartoonish",
+                900,
+                1200,
+                12,
+                41
+            ]
+        },
+        {
+            "id": 13,
+            "type": "SaveVideo",
+            "pos": [
+                1257.5441369501657,
+                -44.12956394419863
+            ],
+            "size": [
+                374,
+                372.34896826553796
+            ],
+            "flags": {},
+            "order": 5,
+            "mode": 0,
+            "inputs": [
+                {
+                    "localized_name": "video",
+                    "name": "video",
+                    "type": "VIDEO",
+                    "link": 26
+                },
+                {
+                    "localized_name": "filename_prefix",
+                    "name": "filename_prefix",
+                    "type": "STRING",
+                    "widget": {
+                        "name": "filename_prefix"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "format",
+                    "name": "format",
+                    "type": "COMBO",
+                    "widget": {
+                        "name": "format"
+                    },
+                    "link": null
+                },
+                {
+                    "localized_name": "codec",
+                    "name": "codec",
+                    "type": "COMBO",
+                    "widget": {
+                        "name": "codec"
+                    },
+                    "link": null
+                }
+            ],
+            "outputs": [],
+            "properties": {},
+            "widgets_values": [
+                "video/ComfyUI",
+                "auto",
+                "auto"
+            ]
+        }
+    ],
+    "links": [
+        [
+            25,
+            4,
+            0,
+            12,
+            0,
+            "IMAGE"
+        ],
+        [
+            26,
+            12,
+            0,
+            13,
+            0,
+            "VIDEO"
+        ],
+        [
+            27,
+            8,
+            0,
+            12,
+            1,
+            "SAMPLING_PARAMS"
+        ],
+        [
+            28,
+            14,
+            0,
+            12,
+            3,
+            "VIDEO_PARAMS"
+        ]
+    ],
+    "groups": [
+        {
+            "id": 1,
+            "title": "Input",
+            "bounding": [
+                213.1706010087147,
+                -313.20095750667554,
+                560.8246471437027,
+                407.4246532472182
+            ],
+            "color": "#3f789e",
+            "font_size": 24,
+            "flags": {}
+        }
+    ],
+    "config": {},
+    "extra": {
+        "workflowRendererVersion": "LG",
+        "ds": {
+            "scale": 1.1469075819486894,
+            "offset": [
+                101.7632950847089,
+                492.29121889347806
+            ]
+        }
+    },
+    "version": 0.4
+}

apps/lvsa-omni-showcase/README.md

@@ -1,87 +0,0 @@
-# LVSA Long-Video Showcase
-
-**Training-free block-sparse attention for long-video diffusion**, running on
-vLLM-Omni via the public plugin entry points (no core changes). LVSA speeds up
-extended-horizon generation and prevents the freeze/loop failure mode dense
-attention exhibits beyond a model's training length — with no fine-tuning and no
-weight changes.
-
-Upstream: **[LongVideoSparseAttention](https://github.com/JiusiServe/LongVideoSparseAttention)**
-(algorithm, all model adapters, full docs).
-
-## Results
-
-Wan 2.1 1.3B, single A100 80 GB, mean over 5 prompts (Dense vs LVSA-FlashInfer):
-
-| Horizon | 1× | 2× | 3× | 4× | 5× | 6× |
-|---|---|---|---|---|---|---|
-| **Speedup** | 0.9× | 1.4× | 1.9× | 2.5× | 3.0× | **3.5×** |
-| **VQeval composite** (Dense→LVSA) | 63→63 | 59→63 | 61→63 | 61→64 | 59→64 | 59→63 |
-
-The speedup **grows with length** (LVSA is the dense regime at 1× by design, so
-the win starts at 2×), and quality *improves* at extension — LVSA's rotating
-keyframes prevent the looping/static output dense produces beyond its horizon
-(VQeval loop-quality +27 to +34 at ≥3×). At long horizons dense **OOMs on 80 GB**
-(HunyuanVideo ≥2×, Wan2.1-14B and Cosmos at 6×) where LVSA still generates.
-
-→ Full sweep (5 models × 6 horizons, VQeval + VBench-Long) and the
-quality-metric methodology:
-**[benchmarks/](https://github.com/JiusiServe/LongVideoSparseAttention/tree/main/benchmarks)**.
-
-## Quickstart
-
-```bash
-# 1. vLLM-Omni 0.22.0 (stable) + matching vLLM
-pip install "vllm==0.22.0"
-pip install --no-build-isolation \
-  "vllm-omni @ git+https://github.com/vllm-project/vllm-omni.git@v0.22.0"
-
-# 2. LVSA core + plugin
-git clone https://github.com/JiusiServe/LongVideoSparseAttention.git
-( cd LongVideoSparseAttention && pip install -e . && pip install -e lvsa-vllm-omni/ )
-
-# 3. this showcase's client dep
-pip install requests
-```
-
-FlashInfer (used by the default `flashinfer` backend) **ships with vLLM** — step 1
-already pulls the pinned `flashinfer-python` + `flashinfer-cubin`, so the fused
-backend works with no extra install. Two optional extras:
-
-```bash
-# (optional) faster startup — prebuilt JIT cache, matched to your CUDA + flashinfer
-CU=cu128   # match your CUDA: cu121 / cu124 / cu128 / …
-FIVER=$(python -c "import importlib.metadata as m; print(m.version('flashinfer-python'))")
-pip install --extra-index-url "https://flashinfer.ai/whl/${CU}" \
-  "flashinfer-jit-cache==${FIVER}+${CU}"
-
-# (fallback) no FlashInfer in your environment? run the SDPA backend instead:
-export LVSA_BACKEND=sdpa
-```
-
-```bash
-# Serve an LVSA-enabled endpoint (Wan 2.1 1.3B at 2× horizon):
-MODEL=/path/to/Wan2.1-T2V-1.3B-Diffusers MODEL_FAMILY=wan FRAMES=161 \
-  bash serve_lvsa.sh
-
-# In another shell — generate:
-python generate.py --frames 161 --height 480 --width 832 \
-  --prompt "A dog running through a sunlit forest." --out wan_lvsa.mp4
-```
-
-`serve_lvsa.sh` sets the LVSA env vars for the chosen `MODEL_FAMILY`
-(`wan` / `hunyuan`). Wan and HunyuanVideo run through the LVSA **attention
-backend** by default — it stays sparse even under sequence-parallel, unlike the
-monkey-patch hooks.
-
-## Configuration
-
-The one knob you must get right is the model's training horizon
-(`LVSA_REFERENCE_LATENT_FRAMES`: Wan = 21, HunyuanVideo = 33); `serve_lvsa.sh`
-sets it per family. For the full environment-variable reference, the multi-GPU
-support matrix, and tuning (`sparsity_scale`, rotating keyframes, non-standard
-resolutions), see the upstream docs:
-
-- **[lvsa-vllm-omni plugin reference](https://github.com/JiusiServe/LongVideoSparseAttention/blob/main/lvsa-vllm-omni/README.md)** — every `LVSA_*` env var
-- **[docs/parallelism.md](https://github.com/JiusiServe/LongVideoSparseAttention/blob/main/docs/parallelism.md)** — TP / Ulysses / HSDP support per model
-- **[docs/tuning.md](https://github.com/JiusiServe/LongVideoSparseAttention/blob/main/docs/tuning.md)** — picking knobs for your model and horizon

apps/lvsa-omni-showcase/generate.py

@@ -1,93 +0,0 @@
-#!/usr/bin/env python
-"""Generate a long video through an LVSA-enabled vLLM-Omni server.
-
-A thin client over the Omni video API — start the server first with
-``serve_lvsa.sh``. The sparse attention runs server-side; this just submits a
-job and saves the result.
-
-Uses the **asynchronous** endpoint (``POST /v1/videos`` → poll
-``GET /v1/videos/{id}`` → download ``GET /v1/videos/{id}/content``) rather than
-``POST /v1/videos/sync``. The sync endpoint is capped by a fixed server-side
-timeout (``VIDEO_SYNC_TIMEOUT_S``, 600 s) and is documented as
-"for benchmark and testing scenarios" — long videos (LVSA's use case) routinely
-exceed it. The async job has no such cap.
-
-  python generate.py --frames 481 --height 480 --width 832 \
-      --prompt "A golden retriever running through a sunlit forest." \
-      --out dog_forest_lvsa.mp4
-"""
-
-import argparse
-import os
-import time
-
-import requests
-
-
-def main():
-    ap = argparse.ArgumentParser()
-    ap.add_argument("--server", default=os.environ.get("OMNI_SERVER", "http://localhost:8000"))
-    ap.add_argument("--prompt", default="A dog running through a sunlit forest.")
-    ap.add_argument("--frames", type=int, default=161)  # Wan 2x horizon
-    ap.add_argument("--height", type=int, default=480)
-    ap.add_argument("--width", type=int, default=832)
-    ap.add_argument("--fps", type=int, default=16)
-    ap.add_argument("--steps", type=int, default=50)
-    ap.add_argument("--guidance", type=float, default=5.0)
-    ap.add_argument("--out", default="lvsa_long_video.mp4")
-    ap.add_argument("--poll-interval", type=float, default=5.0, help="Seconds between status polls.")
-    ap.add_argument(
-        "--max-wait", type=float, default=7200.0, help="Give up if the job hasn't finished after this many seconds."
-    )
-    args = ap.parse_args()
-
-    base = args.server.rstrip("/")
-    form = {
-        "prompt": args.prompt,
-        "size": f"{args.width}x{args.height}",
-        "num_frames": str(args.frames),
-        "fps": str(args.fps),
-        "num_inference_steps": str(args.steps),
-        "guidance_scale": str(args.guidance),
-    }
-
-    # ── 1. submit the async job ──────────────────────────────────────────────
-    print(f"[lvsa-app] POST {base}/v1/videos  {args.width}x{args.height} {args.frames}f steps={args.steps}")
-    resp = requests.post(f"{base}/v1/videos", data=form, timeout=120)
-    resp.raise_for_status()
-    job = resp.json()
-    vid = job["id"]
-    print(f"[lvsa-app] job {vid} ({job.get('status', '?')})")
-
-    # ── 2. poll until completed / failed ─────────────────────────────────────
-    deadline = time.time() + args.max_wait
-    last = None
-    while True:
-        if time.time() > deadline:
-            raise SystemExit(f"[lvsa-app] timed out after {args.max_wait:.0f}s (job {vid} still running)")
-        time.sleep(args.poll_interval)
-        r = requests.get(f"{base}/v1/videos/{vid}", timeout=60)
-        if r.status_code != 200:
-            # retrieve_video returns a non-200 JSON when the job has FAILED
-            raise SystemExit(f"[lvsa-app] job failed: {r.text}")
-        st = r.json()
-        status = st.get("status")
-        progress = st.get("progress", 0)
-        if (status, progress) != last:
-            print(f"[lvsa-app] {status}  {progress}%")
-            last = (status, progress)
-        if status == "completed":
-            break
-        if status == "failed":
-            raise SystemExit(f"[lvsa-app] job failed: {st.get('error')}")
-
-    # ── 3. download the rendered mp4 ─────────────────────────────────────────
-    c = requests.get(f"{base}/v1/videos/{vid}/content", timeout=600)
-    c.raise_for_status()
-    with open(args.out, "wb") as f:
-        f.write(c.content)
-    print(f"[lvsa-app] saved {args.out}")
-
-
-if __name__ == "__main__":
-    main()

apps/lvsa-omni-showcase/serve_lvsa.sh

@@ -1,45 +0,0 @@
-#!/usr/bin/env bash
-# Start an LVSA-enabled vLLM-Omni server for Wan or HunyuanVideo.
-#
-# LVSA is selected as the attention backend via --diffusion-attention-config (see
-# below); the `LVSA_*` env vars only tune it. The `lvsa-vllm-omni` plugin registers
-# itself through vLLM-Omni's `general_plugins` entry points — nothing in vLLM-Omni
-# core is modified. Watch the log for "[LVSA] Geometry detected" to confirm sparse
-# attention is active for your request geometry.
-#
-#   MODEL=/path/to/Wan2.1-T2V-1.3B-Diffusers MODEL_FAMILY=wan     FRAMES=161 bash serve_lvsa.sh
-#   MODEL=/path/to/HunyuanVideo-1.5-...       MODEL_FAMILY=hunyuan FRAMES=193 bash serve_lvsa.sh
-set -euo pipefail
-
-MODEL=${MODEL:?set MODEL=/path/to/Wan-or-HunyuanVideo-Diffusers}
-MODEL_FAMILY=${MODEL_FAMILY:-wan}     # wan | hunyuan
-PORT=${PORT:-8000}
-FRAMES=${FRAMES:-161}                 # target horizon in frames
-VAE_T=${VAE_T:-4}                     # VAE temporal compression (Wan/Hunyuan = 4)
-BACKEND=${LVSA_BACKEND:-flashinfer}   # flashinfer (fused, fastest) | sdpa
-
-T_LAT=$(( (FRAMES - 1) / VAE_T + 1 ))
-
-# Per-family training reference horizon (Wan train ref = 21, Hunyuan = 33).
-case "$MODEL_FAMILY" in
-  wan)     REFERENCE=${REFERENCE:-21} ;;
-  hunyuan) REFERENCE=${REFERENCE:-33} ;;
-  *) echo "MODEL_FAMILY must be 'wan' or 'hunyuan'"; exit 1 ;;
-esac
-export LVSA_BACKEND="$BACKEND"
-export LVSA_TOTAL_LATENT_FRAMES="$T_LAT"
-export LVSA_REFERENCE_LATENT_FRAMES="$REFERENCE"
-export LVSA_ROTATE_KEYFRAMES=${LVSA_ROTATE_KEYFRAMES:-1}
-
-echo "[lvsa-app] serving $MODEL ($MODEL_FAMILY) on :$PORT"
-echo "[lvsa-app]   LVSA backend=$BACKEND  T_lat=$T_LAT (frames=$FRAMES)  reference=$REFERENCE"
-echo "[lvsa-app]   confirm activation by grep '\\[LVSA\\] Geometry detected' in this log"
-
-# vLLM-Omni 0.22 selects the LVSA attention backend per-role via
-# --diffusion-attention-config. The backend is what drives sparse attention
-# here; it engages under tensor-parallel too, unlike the LVSA_*_HOOK monkey-patch
-# path (which falls back to dense under sequence-parallel), so we deliberately do
-# NOT set a hook.
-exec vllm serve "$MODEL" --omni --host 0.0.0.0 --port "$PORT" \
-     --diffusion-attention-config '{"per_role": {"self": {"backend": "LVSA"}}}' \
-     --init-timeout 1800 ${EXTRA_ARGS:-}

benchmarks/README.md

@@ -28,14 +28,6 @@ Benchmarks for GLM-Image performance across HuggingFace baseline, vLLM-Omni offl
 - **Tasks**: text-to-image and image-to-image
 - **Key metrics**: request/image throughput, latency percentiles, optional per-stage pipeline timings
 
-### [LingBot-Video](lingbot_video/README.md) — Dense and MoE Parity
-
-Manual cross-runtime validation for the dense LingBot-Video pipeline and the
-LingBot-Video MoE transformer.
-
-- **Dense pipeline**: decoded-video MAE, MSE, PSNR, latency, and optional steady-state timings
-- **MoE transformer**: bitwise router, sparse-block, shared-expert, and full-transformer parity
-
 ### [Distributed](distributed/omni_connectors/README.md) — RDMA Connector Testing
 
 RDMA environment setup and transfer tests for `MooncakeTransferEngineConnector`, including pytest-based single-node checks and manual cross-node benchmarks.

benchmarks/diffusion/README.md

@@ -32,6 +32,8 @@ python3 benchmarks/diffusion/diffusion_benchmark_serving.py \
 
 - By default, image tasks talk to `http://<host>:<port>/v1/chat/completions`; video tasks talk to `/v1/videos`.
 - If you run the server on another host or port, pass `--base-url` accordingly.
+- To compare diffusion multi-replica serving, restart the server with different
+  `runtime.num_replicas` values and keep the benchmark parameters unchanged.
 
 ## 2. Supported Datasets
 
@@ -93,6 +95,35 @@ You can point to your own trace using `--dataset-path`.
 
 ## 3. Benchmark Parameters
 
+### Multi-replica sweep
+
+For replica scaling, use the same model, prompt count, image/video size, and
+step count for every run. Change only the serving stage config:
+
+```yaml
+runtime:
+  devices: "0,1,2,3"
+  num_replicas: 4
+```
+
+Then run a fixed benchmark and write one JSON file per replica count:
+
+```bash
+python3 benchmarks/diffusion/diffusion_benchmark_serving.py \
+  --base-url http://localhost:8099 \
+  --model Qwen/Qwen-Image \
+  --task t2i \
+  --dataset random \
+  --num-prompts 32 \
+  --max-concurrency 8 \
+  --request-rate inf \
+  --num-inference-steps 10 \
+  --width 512 --height 512 \
+  --output-file diffusion_replica_4.json
+```
+
+Repeat for `num_replicas=1`, `2`, `3`, and `4`, using matching `runtime.devices`.
+
 ### Basic flags
 
 - `--base-url`: Server address; `--endpoint` selects the path appended to this base URL.

benchmarks/diffusion/backends.py

@@ -28,7 +28,6 @@ class RequestFuncInput:
     slo_ms: float | None = None
     extra_body: dict[str, Any] = field(default_factory=dict)
     image_paths: list[str] | None = None
-    video_paths: list[str] | None = None
     request_id: str = field(default_factory=lambda: str(uuid.uuid4()))
     default_bot_task: str | None = DEFAULT_EDITS_BOT_TASK
 
@@ -339,11 +338,6 @@ async def async_request_v1_videos(
         form.add_field(k, str(v))
 
     image_file = None
-    if input.image_paths and input.video_paths:
-        output.error = "Only one of image_paths or video_paths can be provided"
-        output.success = False
-        return output
-
     if input.image_paths and len(input.image_paths) > 0:
         image_path = input.image_paths[0]
         image_file = open(image_path, "rb")
@@ -351,16 +345,7 @@ async def async_request_v1_videos(
             "input_reference",
             image_file,
             filename=os.path.basename(image_path),
-            content_type=_guess_mime_type(image_path),
-        )
-    elif input.video_paths and len(input.video_paths) > 0:
-        video_path = input.video_paths[0]
-        image_file = open(video_path, "rb")
-        form.add_field(
-            "input_reference",
-            image_file,
-            filename=os.path.basename(video_path),
-            content_type=_guess_mime_type(video_path),
+            content_type="application/octet-stream",
         )
 
     job_id = None

benchmarks/diffusion/diffusion_benchmark_serving.py

@@ -73,10 +73,6 @@ Usage:
             {"width":854,"height":480,"num_inference_steps":18,"num_frames":120,"fps":24,"weight":1}
         ]'
 
-    v2v:
-    python3 benchmarks/diffusion/diffusion_benchmark_serving.py \
-        --endpoint /v1/videos --dataset random --task v2v --num-prompts 1 \
-        --height 720 --width 1280 --fps 24 --num-frames 189 --num-inference-steps 35
 
 """
 
@@ -546,17 +542,11 @@ class TraceDataset(BaseDataset):
         if not image_paths:
             single = row.get("image_path")
             image_paths = [single] if single else None
-        video_paths = row.get("video_paths")
-        if not video_paths:
-            single_video = row.get("video_path")
-            video_paths = [single_video] if single_video else None
 
         if not image_paths and self.args.task in ["i2v", "i2i", "ti2v", "ti2i", "it2i"]:
             raise ValueError(
                 f"Task {self.args.task} requires image input, but no image_path or image_paths found in trace row."
             )
-        if not video_paths and self.args.task == "v2v":
-            raise ValueError("Task v2v requires video input, but no video_path or video_paths found in trace row.")
 
         override_w = self.args.width
         override_h = self.args.height
@@ -580,7 +570,6 @@ class TraceDataset(BaseDataset):
             timestamp=timestamp,
             slo_ms=slo_ms,
             image_paths=image_paths,
-            video_paths=video_paths,
             request_id=str(row.get("request_id")) if row.get("request_id") is not None else str(uuid.uuid4()),
         )
 
@@ -600,8 +589,6 @@ class CustomDataset(BaseDataset):
     - seed (optional): Random seed
     - image_paths (optional): List of input image paths for i2i/i2v/ti2i tasks
     - image_urls (optional): List of input image URLs (alternative to image_paths)
-    - video_paths (optional): List of input video paths for v2v tasks
-    - video_urls (optional): List of input video URLs (alternative to video_paths)
 
     Example JSONL for ti2i:
     {"prompt": "Add sunset lighting", "width": 1024, "height": 1024, "image_urls": ["https://example.com/image.jpg"]}
@@ -673,42 +660,6 @@ class CustomDataset(BaseDataset):
 
         return None
 
-    def _resolve_video_paths(self, item: dict) -> list[str] | None:
-        if "video_paths" in item and item["video_paths"]:
-            paths = item["video_paths"]
-            if isinstance(paths, str):
-                paths = [paths]
-
-            valid_paths = []
-            for path in paths:
-                if os.path.exists(path):
-                    valid_paths.append(path)
-                else:
-                    raise ValueError(f"Video file not found: {path}")
-
-            return valid_paths if valid_paths else None
-
-        if "video_urls" in item and item["video_urls"]:
-            urls = item["video_urls"]
-            if isinstance(urls, str):
-                urls = [urls]
-
-            downloaded_paths = []
-            for url in urls:
-                try:
-                    response = requests.get(url, timeout=30)
-                    response.raise_for_status()
-                    suffix = os.path.splitext(url)[1] or ".mp4"
-                    with tempfile.NamedTemporaryFile(delete=False, suffix=suffix) as tmp:
-                        tmp.write(response.content)
-                        downloaded_paths.append(tmp.name)
-                except Exception as e:
-                    raise ValueError(f"Failed to download video from {url}: {e}")
-
-            return downloaded_paths if downloaded_paths else None
-
-        return None
-
     def __len__(self) -> int:
         return len(self.data)
 
@@ -721,22 +672,11 @@ class CustomDataset(BaseDataset):
         height = item.get("height", self.args.height)
         num_inference_steps = item.get("num_inference_steps", self.args.num_inference_steps)
         seed = item.get("seed", self.args.seed)
-        reserved_keys = {
-            "prompt",
-            "width",
-            "height",
-            "num_inference_steps",
-            "seed",
-            "image_paths",
-            "image_urls",
-            "video_paths",
-            "video_urls",
-        }
+        reserved_keys = {"prompt", "width", "height", "num_inference_steps", "seed", "image_paths", "image_urls"}
         extra_body = {k: v for k, v in item.items() if k not in reserved_keys}
 
         # Handle image paths/URLs
         image_paths = self._resolve_image_paths(item)
-        video_paths = self._resolve_video_paths(item)
 
         return RequestFuncInput(
             prompt=prompt,
@@ -744,7 +684,6 @@ class CustomDataset(BaseDataset):
             model=self.model,
             seed=seed,
             image_paths=image_paths,
-            video_paths=video_paths,
             extra_body=extra_body,
             width=width,
             height=height,
@@ -798,10 +737,6 @@ class RandomDataset(BaseDataset):
             self._random_image_path = self._generate_random_image_paths()
         else:
             self._random_image_path = None
-        if self.args.task == "v2v":
-            self._random_video_path = self._generate_random_video_paths()
-        else:
-            self._random_video_path = None
 
     def __len__(self) -> int:
         return self.num_prompts
@@ -821,15 +756,13 @@ class RandomDataset(BaseDataset):
         if self._sampled_requests:
             profile = self._sampled_requests[idx]
             params.update(profile)
-        prompt = str(params.pop("prompt", f"Random prompt {idx} for benchmarking diffusion models"))
         return RequestFuncInput(
-            prompt=prompt,
+            prompt=f"Random prompt {idx} for benchmarking diffusion models",
             api_url=self.api_url,
             model=self.model,
             seed=self.args.seed,
             extra_body=extra_body,
             image_paths=self._random_image_path,
-            video_paths=self._random_video_path,
             **params,
         )
 
@@ -848,35 +781,6 @@ class RandomDataset(BaseDataset):
             image_paths.append(image_path)
         return image_paths
 
-    def _generate_random_video_paths(self) -> list[str]:
-        video_path = os.path.join(tempfile.gettempdir(), "diffusion_benchmark_random_video.mp4")
-        try:
-            import cv2
-
-            profile = self.random_request_config[0] if self.random_request_config else {}
-            width = int(self.args.width or profile.get("width") or 1280)
-            height = int(self.args.height or profile.get("height") or 720)
-            num_frames = int(self.args.num_frames or profile.get("num_frames") or 16)
-            fps = float(self.args.fps or profile.get("fps") or 8)
-            writer = cv2.VideoWriter(video_path, cv2.VideoWriter_fourcc(*"mp4v"), fps, (width, height))
-            if not writer.isOpened():
-                raise RuntimeError("cv2.VideoWriter failed to open")
-            for frame_idx in range(num_frames):
-                frame = np.zeros((height, width, 3), dtype=np.uint8)
-                x = int((frame_idx / max(num_frames - 1, 1)) * max(width - 96, 1))
-                y = height // 2
-                frame[:, :, 0] = 24
-                frame[:, :, 1] = 32
-                frame[:, :, 2] = 48
-                frame[max(y - 36, 0) : min(y + 36, height), x : min(x + 96, width), :] = (64, 128, 220)
-                writer.write(frame)
-            writer.release()
-            return [video_path]
-        except Exception as e:
-            raise RuntimeError(
-                "Failed to generate synthetic v2v input video. Install opencv-python or provide a custom dataset."
-            ) from e
-
 
 def _compute_expected_latency_ms_from_base(req: RequestFuncInput, args, base_time_ms: float | None) -> float | None:
     """Compute expected execution time (ms) based on a base per-step-per-frame unit time.
@@ -1127,14 +1031,6 @@ def _save_generated_outputs(
         if not out.success or not out.response_body:
             continue
 
-        if isinstance(out.response_body, bytes):
-            fname = f"req_{idx:04d}.mp4"
-            fpath = os.path.join(save_dir, fname)
-            with open(fpath, "wb") as f:
-                f.write(out.response_body)
-            saved += 1
-            continue
-
         media_urls: list[str] = []
 
         # Chat-completions style: choices[*].message.content[*].image_url.url
@@ -1208,7 +1104,7 @@ def wait_for_service(base_url: str, timeout: int = 120) -> None:
 
 
 def _default_endpoint_for_task(task: str) -> str:
-    if task in {"t2v", "i2v", "ti2v", "v2v"}:
+    if task in {"t2v", "i2v", "ti2v"}:
         return "/v1/videos"
     if task in {"i2i", "ti2i", "it2i"}:
         return "/v1/images/edits"
@@ -1222,7 +1118,7 @@ async def benchmark(args):
     if args.base_url is None:
         args.base_url = f"http://{args.host}:{args.port}"
 
-    VIDEO_TASKS = {"t2v", "i2v", "ti2v", "v2v"}
+    VIDEO_TASKS = {"t2v", "i2v", "ti2v"}
     IMAGE_TASKS = {"t2i", "i2i", "ti2i", "it2i"}
 
     if args.task in VIDEO_TASKS:
@@ -1273,10 +1169,6 @@ async def benchmark(args):
     if args.return_stage_metrics and args.endpoint in _STAGE_METRICS_ENDPOINTS:
         for req in requests_list:
             req.extra_body.setdefault(_RETURN_STAGE_METRICS_FIELD, True)
-    if args.extra_body:
-        extra_body = json.loads(args.extra_body)
-        for req in requests_list:
-            req.extra_body.update(extra_body)
 
     if args.endpoint == "/v1/images/edits":
         for req in requests_list:
@@ -1428,7 +1320,7 @@ if __name__ == "__main__":
         "--task",
         type=str,
         default="t2v",
-        choices=["t2v", "i2v", "ti2v", "v2v", "ti2i", "i2i", "it2i", "t2i"],
+        choices=["t2v", "i2v", "ti2v", "ti2i", "i2i", "it2i", "t2i"],
         help="Task type.",
     )
     parser.add_argument(
@@ -1543,12 +1435,6 @@ if __name__ == "__main__":
             '{"width":768,"height":768,"num_inference_steps":20,"weight":0.85}]'
         ),
     )
-    parser.add_argument(
-        "--extra-body",
-        type=str,
-        default=None,
-        help="JSON object merged into every generated request body/form.",
-    )
     parser.add_argument(
         "--num-input-images",
         type=int,

benchmarks/fish-speech/bench_speaker_cache.py

@@ -201,7 +201,6 @@ async def main():
             "input": prompt,
             "voice": "default",
             "stream": True,
-            "stream_format": "audio",
             "response_format": "pcm",
             "ref_audio": ref_audio_b64,
             "ref_text": args.ref_text,
@@ -241,7 +240,6 @@ async def main():
             "input": prompt,
             "voice": args.voice_name,
             "stream": True,
-            "stream_format": "audio",
             "response_format": "pcm",
             "ref_text": args.ref_text,
             "max_new_tokens": 2048,

benchmarks/lingbot_video/README.md

@@ -1,62 +0,0 @@
-# LingBot-Video Benchmarks
-
-These manual benchmarks compare the upstream LingBot-Video implementation with
-vLLM-Omni. Run all commands from the repository root.
-
-| Benchmark | Script | Purpose |
-|-----------|--------|---------|
-| Dense pipeline parity | `dense_pipeline_parity.py` | Compare generated videos and report MAE, MSE, PSNR, latency, and optional steady-state timings |
-| MoE numerical parity | `moe_transformer_parity.py` | Check bitwise parity for the sparse MoE block and the full MoE transformer |
-
-Both benchmarks require a local LingBot-Video checkout and local model files.
-They do not run in CI.
-
-## Dense pipeline parity
-
-The dense benchmark launches the upstream Diffusers inference script and the
-vLLM-Omni offline example with identical prompts and sampling parameters. It
-then compares the decoded MP4 frames.
-
-```bash
-CUDA_VISIBLE_DEVICES=0 \
-python benchmarks/lingbot_video/dense_pipeline_parity.py \
-  --model /path/to/lingbot-video-dense-1.3b \
-  --official-repo /path/to/lingbot-video \
-  --output-dir /tmp/lingbot_dense_parity
-```
-
-Use `--runs N` to additionally measure repeated in-process vLLM-Omni
-requests. The video comparison reports MAE, MSE, and PSNR; it is an end-to-end
-pipeline diagnostic rather than a bitwise transformer check.
-
-## MoE numerical parity
-
-The lightweight block comparison uses deterministic weights and inputs to
-cover correction-bias routing, group-limited top-k, routed experts, FP32
-scatter-weighted restore, padding masks, and the shared expert.
-
-```bash
-CUDA_VISIBLE_DEVICES=0 \
-python benchmarks/lingbot_video/moe_transformer_parity.py \
-  --scope block \
-  --official-repo /path/to/lingbot-video \
-  --output-json /tmp/lingbot_moe_block_parity.json
-```
-
-The full comparison loads the upstream and vLLM-Omni 30B transformers
-sequentially and compares their output tensors. It requires a CUDA device with
-`torch._grouped_mm` and enough memory for one 30B BF16 transformer.
-
-```bash
-CUDA_VISIBLE_DEVICES=0 \
-python benchmarks/lingbot_video/moe_transformer_parity.py \
-  --scope transformer \
-  --official-repo /path/to/lingbot-video \
-  --model /path/to/lingbot-video-moe-30b-a3b \
-  --output-json /tmp/lingbot_moe_transformer_parity.json
-```
-
-For the correctness oracle, the upstream transformer uses
-`diffusers:_native_math` and vLLM-Omni uses
-`TORCH_SDPA + SDPBackend.MATH`. The command exits with a nonzero status unless
-the selected comparisons are bitwise equal.

benchmarks/lingbot_video/dense_pipeline_parity.py

@@ -1,217 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-
-from __future__ import annotations
-
-import argparse
-import json
-import re
-import statistics
-import subprocess
-import sys
-import time
-from pathlib import Path
-from typing import Any
-
-import imageio.v3 as iio
-import numpy as np
-import torch
-
-from vllm_omni.diffusion.data import DiffusionParallelConfig
-from vllm_omni.entrypoints.omni import Omni
-from vllm_omni.inputs.data import OmniDiffusionSamplingParams
-from vllm_omni.platforms import current_omni_platform
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="LingBot dense parity and timing harness.")
-    parser.add_argument("--model", default="/home/models/lingbot-video-dense-1.3b")
-    parser.add_argument("--official-repo", default="/tmp/lingbot-video")
-    parser.add_argument("--output-dir", default="/tmp/lingbot_dense_parity")
-    parser.add_argument("--prompt", default="a robotic arm picks up a red block")
-    parser.add_argument("--height", type=int, default=192)
-    parser.add_argument("--width", type=int, default=320)
-    parser.add_argument("--num-frames", type=int, default=9)
-    parser.add_argument("--steps", type=int, default=2)
-    parser.add_argument("--guidance-scale", type=float, default=3.0)
-    parser.add_argument("--shift", type=float, default=3.0)
-    parser.add_argument("--seed", type=int, default=42)
-    parser.add_argument("--fps", type=int, default=24)
-    parser.add_argument("--runs", type=int, default=0, help="Optional in-process steady-state native runs.")
-    return parser.parse_args()
-
-
-def _run(cmd: list[str], cwd: Path | None = None) -> tuple[float, str]:
-    start = time.perf_counter()
-    proc = subprocess.run(cmd, cwd=cwd, text=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
-    elapsed = time.perf_counter() - start
-    if proc.returncode != 0:
-        print(proc.stdout)
-        raise subprocess.CalledProcessError(proc.returncode, proc.args, output=proc.stdout)
-    return elapsed, proc.stdout
-
-
-def _read_video(path: Path) -> np.ndarray:
-    frames = iio.imread(path, plugin="pyav")
-    if frames.dtype != np.float32:
-        frames = frames.astype(np.float32) / 255.0
-    return frames
-
-
-def _compare(a: np.ndarray, b: np.ndarray) -> dict[str, Any]:
-    if a.shape != b.shape:
-        return {"shape_a": list(a.shape), "shape_b": list(b.shape), "shape_match": False}
-    diff = a - b
-    mse = float(np.mean(diff * diff))
-    mae = float(np.mean(np.abs(diff)))
-    psnr = float("inf") if mse == 0 else float(20.0 * np.log10(1.0 / np.sqrt(mse)))
-    return {
-        "shape": list(a.shape),
-        "shape_match": True,
-        "mae": mae,
-        "mse": mse,
-        "psnr": psnr,
-    }
-
-
-def _extract_total_generation_seconds(log: str) -> float | None:
-    match = re.search(r"Total generation time:\s*([0-9.]+)\s*seconds", log)
-    return float(match.group(1)) if match else None
-
-
-def _peak_memory_mb(output: Any) -> float:
-    result = output[0] if isinstance(output, list) and output else output
-    return float(getattr(result, "peak_memory_mb", 0.0) or 0.0)
-
-
-def _benchmark_native_requests(args: argparse.Namespace) -> dict[str, Any]:
-    load_start = time.perf_counter()
-    omni = Omni(
-        model=args.model,
-        model_class_name="LingBotVideoPipeline",
-        flow_shift=args.shift,
-        parallel_config=DiffusionParallelConfig(),
-    )
-    load_seconds = time.perf_counter() - load_start
-    prompt = {"prompt": args.prompt}
-    latencies: list[float] = []
-    peak_mb = 0.0
-
-    for _ in range(args.runs):
-        generator = torch.Generator(device=current_omni_platform.device_type).manual_seed(args.seed)
-        sampling_params = OmniDiffusionSamplingParams(
-            height=args.height,
-            width=args.width,
-            num_frames=args.num_frames,
-            num_inference_steps=args.steps,
-            guidance_scale=args.guidance_scale,
-            generator=generator,
-            output_type="pt",
-            extra_args={"shift": args.shift},
-        )
-        start = time.perf_counter()
-        output = omni.generate(prompt, sampling_params)
-        latencies.append(time.perf_counter() - start)
-        peak_mb = max(peak_mb, _peak_memory_mb(output))
-        del output
-    omni.close()
-
-    return {
-        "load_seconds": load_seconds,
-        "run_seconds": latencies,
-        "mean_seconds": statistics.mean(latencies) if latencies else None,
-        "median_seconds": statistics.median(latencies) if latencies else None,
-        "min_seconds": min(latencies) if latencies else None,
-        "max_seconds": max(latencies) if latencies else None,
-        "peak_mb": peak_mb,
-    }
-
-
-def main() -> None:
-    args = parse_args()
-    output_dir = Path(args.output_dir)
-    output_dir.mkdir(parents=True, exist_ok=True)
-    official_out = output_dir / "official_diffusers.mp4"
-    native_out = output_dir / "vllm_omni_native.mp4"
-
-    official_cmd = [
-        sys.executable,
-        str(Path(args.official_repo) / "scripts" / "inference.py"),
-        "--backend",
-        "diffusers",
-        "--model_dir",
-        args.model,
-        "--mode",
-        "t2v",
-        "--prompt",
-        args.prompt,
-        "--output",
-        str(official_out),
-        "--height",
-        str(args.height),
-        "--width",
-        str(args.width),
-        "--num_frames",
-        str(args.num_frames),
-        "--steps",
-        str(args.steps),
-        "--guidance_scale",
-        str(args.guidance_scale),
-        "--shift",
-        str(args.shift),
-        "--seed",
-        str(args.seed),
-        "--fps",
-        str(args.fps),
-        "--quiet_progress",
-    ]
-    native_cmd = [
-        sys.executable,
-        "examples/offline_inference/text_to_video/text_to_video_lingbot.py",
-        "--model",
-        args.model,
-        "--prompt",
-        args.prompt,
-        "--output",
-        str(native_out),
-        "--height",
-        str(args.height),
-        "--width",
-        str(args.width),
-        "--num-frames",
-        str(args.num_frames),
-        "--num-inference-steps",
-        str(args.steps),
-        "--guidance-scale",
-        str(args.guidance_scale),
-        "--flow-shift",
-        str(args.shift),
-        "--seed",
-        str(args.seed),
-        "--fps",
-        str(args.fps),
-    ]
-
-    official_time, official_log = _run(official_cmd, cwd=Path(args.official_repo))
-    native_time, native_log = _run(native_cmd, cwd=Path.cwd())
-
-    result = {
-        "official_seconds": official_time,
-        "native_process_seconds": native_time,
-        "native_request_seconds": _extract_total_generation_seconds(native_log),
-        "official_output": str(official_out),
-        "native_output": str(native_out),
-        "accuracy": _compare(_read_video(official_out), _read_video(native_out)),
-        "official_log_tail": official_log.splitlines()[-20:],
-        "native_log_tail": native_log.splitlines()[-20:],
-    }
-    if args.runs > 0:
-        result["native_steady_state"] = _benchmark_native_requests(args)
-
-    result_path = output_dir / "comparison.json"
-    result_path.write_text(json.dumps(result, indent=2, sort_keys=True) + "\n")
-    print(json.dumps(result, indent=2, sort_keys=True))
-
-
-if __name__ == "__main__":
-    main()

benchmarks/lingbot_video/moe_transformer_parity.py

@@ -1,399 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-
-"""Reproduce LingBot-Video MoE block and transformer numerical parity."""
-
-from __future__ import annotations
-
-import argparse
-import gc
-import importlib
-import json
-import os
-import sys
-import time
-from pathlib import Path
-from typing import Any
-
-import torch
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description=__doc__)
-    parser.add_argument(
-        "--scope",
-        choices=("block", "transformer", "all"),
-        default="block",
-        help="Run the lightweight sparse-block check, the real transformer check, or both.",
-    )
-    parser.add_argument(
-        "--official-repo",
-        type=Path,
-        required=True,
-        help="Local checkout of https://github.com/Robbyant/lingbot-video.",
-    )
-    parser.add_argument(
-        "--model",
-        default=None,
-        help="Local MoE checkpoint path or a checkpoint already present in the Hugging Face cache.",
-    )
-    parser.add_argument("--transformer-subfolder", default="transformer")
-    parser.add_argument("--output-json", type=Path, default=None)
-    parser.add_argument("--device", default="cuda")
-    parser.add_argument("--seed", type=int, default=42)
-    parser.add_argument("--num-frames", type=int, default=1)
-    parser.add_argument("--height", type=int, default=8)
-    parser.add_argument("--width", type=int, default=8)
-    parser.add_argument("--text-length", type=int, default=16)
-    return parser.parse_args()
-
-
-def _load_official_module(official_repo: Path):
-    transformer_file = official_repo / "lingbot_video" / "transformer_lingbot_video.py"
-    if not transformer_file.is_file():
-        raise FileNotFoundError(f"Upstream transformer not found: {transformer_file}")
-    sys.path.insert(0, str(official_repo))
-    module = importlib.import_module("lingbot_video.transformer_lingbot_video")
-    module_path = Path(module.__file__).resolve()
-    if official_repo.resolve() not in module_path.parents:
-        raise RuntimeError(f"Imported LingBot module from {module_path}, not {official_repo.resolve()}")
-    return module
-
-
-def _load_native_module():
-    return importlib.import_module("vllm_omni.diffusion.models.lingbot_video.lingbot_video_transformer")
-
-
-def _tensor_metrics(actual: torch.Tensor, expected: torch.Tensor) -> dict[str, Any]:
-    actual_float = actual.float()
-    expected_float = expected.float()
-    diff = (actual_float - expected_float).abs()
-    denominator = actual_float.norm() * expected_float.norm()
-    cosine = (
-        float(torch.dot(actual_float.flatten(), expected_float.flatten()) / denominator)
-        if float(denominator) > 0
-        else 1.0
-    )
-    return {
-        "shape": list(actual.shape),
-        "equal": bool(torch.equal(actual, expected)),
-        "max_abs": float(diff.max()) if diff.numel() else 0.0,
-        "mean_abs": float(diff.mean()) if diff.numel() else 0.0,
-        "rmse": float(torch.sqrt(torch.mean(diff.square()))) if diff.numel() else 0.0,
-        "cosine": cosine,
-        "finite": bool(torch.isfinite(actual).all() and torch.isfinite(expected).all()),
-    }
-
-
-def _initialize_block(module: torch.nn.Module, seed: int) -> None:
-    generator = torch.Generator(device="cpu").manual_seed(seed)
-    with torch.no_grad():
-        for parameter in module.parameters():
-            values = torch.randn(parameter.shape, generator=generator, dtype=torch.float32)
-            parameter.copy_((values * 0.02).to(parameter.dtype))
-        module.router.e_score_correction_bias.copy_(torch.tensor([0.9, 0.8, 1.0, 0.0, 0.7, 0.6, 0.5, 0.4]))
-
-
-def _apply_padding(
-    scores: torch.Tensor,
-    padding_mask: torch.Tensor,
-    route_scale: float,
-) -> torch.Tensor:
-    scores = scores * padding_mask.unsqueeze(-1).to(scores.dtype)
-    scores = scores / (scores.sum(dim=-1, keepdim=True) + 1e-9)
-    return scores * route_scale
-
-
-def _run_block_parity(
-    official_module,
-    native_module,
-    *,
-    device: torch.device,
-    seed: int,
-) -> dict[str, Any]:
-    if not hasattr(torch, "_grouped_mm"):
-        raise RuntimeError("The block parity check requires torch._grouped_mm.")
-
-    common = {
-        "hidden_size": 16,
-        "num_experts": 8,
-        "top_k": 2,
-        "moe_intermediate_size": 8,
-        "score_func": "sigmoid",
-        "norm_topk_prob": True,
-        "n_group": 4,
-        "topk_group": 1,
-        "routed_scaling_factor": 1.5,
-        "n_shared_experts": 1,
-    }
-    official = official_module.LingBotVideoSparseMoeBlock(
-        intermediate_size=32,
-        **common,
-    )
-    native = native_module.LingBotVideoSparseMoeBlock(**common)
-    _initialize_block(official, seed)
-    native.load_state_dict(official.state_dict(), strict=True)
-    official = official.to(device=device, dtype=torch.bfloat16).eval()
-    native = native.to(device=device, dtype=torch.bfloat16).eval()
-
-    generator = torch.Generator(device=device).manual_seed(seed)
-    hidden_states = torch.randn(
-        2,
-        5,
-        common["hidden_size"],
-        generator=generator,
-        device=device,
-        dtype=torch.bfloat16,
-    )
-    padding_mask = torch.tensor(
-        [1, 1, 1, 1, 1, 1, 1, 1, 0, 0],
-        device=device,
-        dtype=torch.float32,
-    )
-    tokens = hidden_states.reshape(-1, common["hidden_size"])
-
-    with torch.inference_mode():
-        official_router = official.router(tokens)
-        official_indices, official_scores = official_router[:2]
-        native_indices, native_scores = native.router(tokens)
-
-        official_scores = _apply_padding(
-            official_scores,
-            padding_mask,
-            official.router.route_scale,
-        )
-        native_scores = _apply_padding(
-            native_scores,
-            padding_mask,
-            native.router.route_scale,
-        )
-        official_routed = official._run_selected_experts(
-            tokens,
-            official_scores,
-            official_indices,
-        )
-        native_routed = native._run_selected_experts(
-            tokens,
-            native_scores,
-            native_indices,
-        )
-        official_shared = official.shared_experts(hidden_states)
-        native_shared = native.shared_experts(hidden_states)
-        official_output = official(hidden_states, padding_mask=padding_mask)
-        native_output = native(hidden_states, padding_mask=padding_mask)
-
-    result = {
-        "router_indices_equal": bool(torch.equal(native_indices, official_indices)),
-        "router_scores": _tensor_metrics(native_scores, official_scores),
-        "routed_output": _tensor_metrics(native_routed, official_routed),
-        "shared_output": _tensor_metrics(native_shared, official_shared),
-        "final_output": _tensor_metrics(native_output, official_output),
-    }
-    result["exact"] = bool(
-        result["router_indices_equal"]
-        and result["router_scores"]["equal"]
-        and result["routed_output"]["equal"]
-        and result["shared_output"]["equal"]
-        and result["final_output"]["equal"]
-    )
-    return result
-
-
-def _release_cuda() -> None:
-    gc.collect()
-    torch.accelerator.empty_cache()
-    torch.accelerator.synchronize()
-
-
-def _make_transformer_inputs(
-    config,
-    args: argparse.Namespace,
-) -> dict[str, torch.Tensor]:
-    generator = torch.Generator(device="cpu").manual_seed(args.seed)
-    hidden_states = torch.randn(
-        1,
-        int(config.in_channels),
-        args.num_frames,
-        args.height,
-        args.width,
-        generator=generator,
-        dtype=torch.float32,
-    ).to(torch.bfloat16)
-    encoder_hidden_states = torch.randn(
-        1,
-        args.text_length,
-        int(config.text_dim),
-        generator=generator,
-        dtype=torch.float32,
-    ).to(torch.bfloat16)
-    return {
-        "hidden_states": hidden_states,
-        "timestep": torch.tensor([500.0], dtype=torch.float32),
-        "encoder_hidden_states": encoder_hidden_states,
-        "encoder_attention_mask": torch.ones(1, args.text_length, dtype=torch.long),
-    }
-
-
-def _load_transformer(
-    transformer_cls,
-    args: argparse.Namespace,
-    device: torch.device,
-):
-    start = time.perf_counter()
-    model = transformer_cls.from_pretrained(
-        args.model,
-        subfolder=args.transformer_subfolder,
-        torch_dtype=torch.bfloat16,
-        local_files_only=True,
-        low_cpu_mem_usage=True,
-    )
-    model = model.to(device=device, dtype=torch.bfloat16).eval()
-    torch.accelerator.synchronize()
-    return model, time.perf_counter() - start
-
-
-def _forward_transformer(
-    model,
-    inputs: dict[str, torch.Tensor],
-    device: torch.device,
-    attention_context,
-) -> tuple[torch.Tensor, float, float]:
-    torch.accelerator.reset_peak_memory_stats(device)
-    gpu_inputs = {name: value.to(device) for name, value in inputs.items()}
-    start = time.perf_counter()
-    with torch.inference_mode(), attention_context:
-        output = model(**gpu_inputs, return_dict=False)[0]
-    torch.accelerator.synchronize()
-    elapsed = time.perf_counter() - start
-    peak_gib = torch.accelerator.max_memory_reserved(device) / (1024**3)
-    return output.cpu(), elapsed, peak_gib
-
-
-def _run_transformer_parity(
-    official_module,
-    native_module,
-    *,
-    args: argparse.Namespace,
-    device: torch.device,
-) -> dict[str, Any]:
-    if args.model is None:
-        raise ValueError("--model is required for transformer parity.")
-
-    from diffusers.models.attention_dispatch import attention_backend
-    from torch.nn.attention import SDPBackend, sdpa_kernel
-
-    from vllm_omni.diffusion.config import set_current_diffusion_config
-    from vllm_omni.diffusion.data import AttentionConfig, AttentionSpec, OmniDiffusionConfig
-
-    official, official_load = _load_transformer(
-        official_module.LingBotVideoTransformer3DModel,
-        args,
-        device,
-    )
-    inputs = _make_transformer_inputs(official.config, args)
-    official_output, official_forward, official_peak = _forward_transformer(
-        official,
-        inputs,
-        device,
-        attention_backend("_native_math"),
-    )
-    del official
-    _release_cuda()
-
-    native_config = OmniDiffusionConfig(
-        diffusion_attention_config=AttentionConfig(
-            default=AttentionSpec(backend="TORCH_SDPA"),
-        ),
-    )
-    with set_current_diffusion_config(native_config):
-        native, native_load = _load_transformer(
-            native_module.LingBotVideoTransformer3DModel,
-            args,
-            device,
-        )
-    native_backend_prefs = {block.attn.attn.backend_pref for block in native.blocks}
-    if native_backend_prefs != {"TORCH_SDPA"}:
-        raise RuntimeError(
-            f"Failed to force the native transformer to TORCH_SDPA: resolved preferences were {native_backend_prefs}."
-        )
-    native_output, native_forward, native_peak = _forward_transformer(
-        native,
-        inputs,
-        device,
-        sdpa_kernel(SDPBackend.MATH),
-    )
-    del native
-    _release_cuda()
-
-    metrics = _tensor_metrics(native_output, official_output)
-    return {
-        "backends": {
-            "official": "diffusers:_native_math",
-            "native": "TORCH_SDPA+SDPBackend.MATH",
-        },
-        "official": {
-            "load_seconds": official_load,
-            "forward_seconds": official_forward,
-            "peak_reserved_gib": official_peak,
-        },
-        "native": {
-            "load_seconds": native_load,
-            "forward_seconds": native_forward,
-            "peak_reserved_gib": native_peak,
-        },
-        "output": metrics,
-        "exact": metrics["equal"],
-    }
-
-
-def main() -> int:
-    args = parse_args()
-    device = torch.device(args.device)
-    if device.type != "cuda" or not torch.cuda.is_available():
-        raise RuntimeError("LingBot MoE parity requires an available CUDA device.")
-
-    os.environ["DIFFUSERS_ATTN_BACKEND"] = "_native_math"
-    os.environ["DIFFUSION_ATTENTION_BACKEND"] = "TORCH_SDPA"
-    os.environ["LINGBOT_MOE_EXPERT_BACKEND"] = "grouped_mm"
-    os.environ["LINGBOT_MOE_PAD_BACKEND"] = "loop"
-    os.environ["LINGBOT_MOE_REORDER_BACKEND"] = "sort"
-    os.environ["LINGBOT_MOE_RESTORE_BACKEND"] = "scatter"
-
-    official_module = _load_official_module(args.official_repo)
-    native_module = _load_native_module()
-
-    result: dict[str, Any] = {
-        "settings": {
-            "scope": args.scope,
-            "seed": args.seed,
-            "device": str(device),
-            "official_repo": str(args.official_repo.resolve()),
-            "model": args.model,
-        }
-    }
-    if args.scope in {"block", "all"}:
-        result["block"] = _run_block_parity(
-            official_module,
-            native_module,
-            device=device,
-            seed=args.seed,
-        )
-    if args.scope in {"transformer", "all"}:
-        result["transformer"] = _run_transformer_parity(
-            official_module,
-            native_module,
-            args=args,
-            device=device,
-        )
-
-    result["exact"] = all(section["exact"] for name, section in result.items() if name in {"block", "transformer"})
-    payload = json.dumps(result, indent=2, sort_keys=True)
-    print(payload)
-    if args.output_json is not None:
-        args.output_json.parent.mkdir(parents=True, exist_ok=True)
-        args.output_json.write_text(payload + "\n", encoding="utf-8")
-    return 0 if result["exact"] else 1
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())

benchmarks/tts/model_configs.yaml

@@ -80,17 +80,6 @@ models:
     task_extra_body:
       voice_clone: {}
 
-  # MOSS-TTS-Local-Transformer-v1.5: Qwen3 backbone + 1-layer GPT2-style local
-  # depth transformer (n_vq=12), paired with the 48 kHz stereo
-  # MOSS-Audio-Tokenizer-v2 codec. Same voice-clone request shape as the
-  # other MOSS-TTS variants.
-  OpenMOSS-Team/MOSS-TTS-Local-Transformer-v1.5:
-    supported_tasks: [voice_clone]
-    backend: openai-audio-speech
-    endpoint: /v1/audio/speech
-    task_extra_body:
-      voice_clone: {}
-
   # MOSS-TTSD synthesizes [S1]/[S2] dialogue. The seed-tts voice_clone dataset
   # has single-speaker rows; bench piggybacks on it (both speakers share the
   # same ref_audio). For real dialogue evaluation use the ttsd dataset.

docker/Dockerfile.ci

@@ -1,26 +1,12 @@
-# Target vllm version: v0.26.0 (releases/v0.26.0 branch).
-# Keep VLLM_BASE_TAG aligned with the target vLLM release on every rebase.
+# Target vllm version: v0.21.0 (releases/v0.21.0 branch).
+# The pinned VLLM_PRECOMPILED_WHEEL_COMMIT and VLLM_BASE_TAG below must
+# be updated on every rebase to match the new upstream vllm commit/tag.
 ARG VLLM_BASE_IMAGE=vllm/vllm-openai
-ARG VLLM_BASE_TAG=v0.26.0
-
-# We use registry-based layer caching for the dependencies,
-# where the tag is determined based on the hash of the contents
-# of files like pyproject.toml, etc. The deps stage should have
-# everything except the actual project dir, since that should not
-# change much across branches.
-FROM ${VLLM_BASE_IMAGE}:${VLLM_BASE_TAG} AS deps
-# Keep APP_DIR outside /workspace so k8s Buildkite workspace mounts cannot
-# shadow the editable install target (#5364). Docker WORKDIR creates /opt if
-# needed.
-ARG APP_DIR=/opt/vllm-omni
+ARG VLLM_BASE_TAG=v0.23.0
+FROM ${VLLM_BASE_IMAGE}:${VLLM_BASE_TAG}
+ARG APP_DIR=/workspace/vllm-omni
 WORKDIR ${APP_DIR}
-
-RUN ln -sf /usr/bin/python3 /usr/bin/python
-
-# Placeholder package tree so editable install can register APP_DIR.
-# The final stage COPY --link fills this directory with the real sources;
-# the editable .pth / egg-link keeps resolving imports there — no PYTHONPATH.
-RUN mkdir -p vllm_omni && touch vllm_omni/__init__.py
+COPY . .
 
 # Install system dependencies
 RUN apt-get update && \
@@ -28,30 +14,9 @@ RUN apt-get update && \
     apt-get clean && \
     rm -rf /var/lib/apt/lists/*
 
-# Copy dependency related things first and install deps/
-# We set the version str to 0.0.0 since it doesn't matter
-# for CI
-COPY pyproject.toml setup.py ./
-COPY requirements/ requirements/
-ENV SETUPTOOLS_SCM_PRETEND_VERSION=0.0.0
-# Editable install: deps + project metadata are cached in this layer, while
-# the package itself is linked to APP_DIR (populated later by COPY --link).
-# Do NOT use a non-editable install here — that copies an empty stub into
-# site-packages and breaks ``import vllm_omni.entrypoints``.
-RUN uv pip install --system -e ".[dev]"
+RUN uv pip install --system ".[dev]"
 
-# The final stage just copies the working dir on top of the
-# dependencies image. This uses --link so that we can skip
-# the image pull in Buildkite when we have a cache hit.
-#
-# IMPORTANT: Do not make any modifications that depend on the
-# file system after the --link copy, otherwise we'd need to pull
-# the heavy base image to build it. Editable install above already
-# registered APP_DIR, so we must not re-pip-install after COPY.
-FROM deps
-ARG APP_DIR=/opt/vllm-omni
-WORKDIR ${APP_DIR}
 
-COPY --link . .
+RUN ln -sf /usr/bin/python3 /usr/bin/python
 
 ENTRYPOINT []

docker/Dockerfile.ci.dockerignore

@@ -1,37 +0,0 @@
-# NOTE: this is largely taken from vLLM
-
-# Exclude .git for CI builds — version is set via SETUPTOOLS_SCM_PRETEND_VERSION
-.git
-.venv
-/build
-dist
-
-# Byte-compiled / optimized / DLL files
-__pycache__/
-*.py[cod]
-*$py.class
-
-.mypy_cache
-
-# Distribution / packaging
-.Python
-/build/
-cmake-build-*/
-CMakeUserPresets.json
-develop-eggs/
-/dist/
-downloads/
-eggs/
-.eggs/
-lib/
-lib64/
-parts/
-sdist/
-var/
-wheels/
-share/python-wheels/
-*.egg-info/
-.installed.cfg
-*.egg
-MANIFEST
-rust/target/

docker/Dockerfile.cuda

@@ -1,4 +1,4 @@
-ARG BASE_IMAGE=vllm/vllm-openai:v0.26.0
+ARG BASE_IMAGE=vllm/vllm-openai:v0.23.0
 FROM ${BASE_IMAGE}
 
 ARG COMMON_WORKDIR=/app

docker/Dockerfile.npu

@@ -1,5 +1,5 @@
-ARG VLLM_ASCEND_IMAGE=quay.io/atlas-ci/vllm-ascend
-ARG VLLM_ASCEND_TAG=v0.26.0
+ARG VLLM_ASCEND_IMAGE=quay.io/ascend/vllm-ascend
+ARG VLLM_ASCEND_TAG=v0.18.0rc1
 FROM ${VLLM_ASCEND_IMAGE}:${VLLM_ASCEND_TAG}
 
 # WORKDIR /vllm-workspace/vllm
@@ -26,10 +26,6 @@ RUN export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
     python3 -m pip install -v -e /vllm-workspace/vllm-omni/ --no-build-isolation
 
-RUN python3 -m pip install -v --no-deps \
-      "stepaudio2-minicpmo>=0.1.1" \
-      "step-audio2>=1.0.0"
-
 ENV VLLM_WORKER_MULTIPROC_METHOD=spawn
 
 ENTRYPOINT []

docker/Dockerfile.npu.a3

@@ -1,5 +1,5 @@
-ARG VLLM_ASCEND_IMAGE=quay.io/atlas-ci/vllm-ascend
-ARG VLLM_ASCEND_TAG=v0.26.0-a3
+ARG VLLM_ASCEND_IMAGE=quay.io/ascend/vllm-ascend
+ARG VLLM_ASCEND_TAG=v0.18.0rc1-a3
 FROM ${VLLM_ASCEND_IMAGE}:${VLLM_ASCEND_TAG}
 
 # WORKDIR /vllm-workspace/vllm
@@ -26,10 +26,6 @@ RUN export PIP_EXTRA_INDEX_URL=https://mirrors.huaweicloud.com/ascend/repos/pypi
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
     python3 -m pip install -v -e /vllm-workspace/vllm-omni/ --no-build-isolation
 
-RUN python3 -m pip install -v --no-deps \
-      "stepaudio2-minicpmo>=0.1.1" \
-      "step-audio2>=1.0.0"
-
 ENV VLLM_WORKER_MULTIPROC_METHOD=spawn
 
 ENTRYPOINT []

docker/Dockerfile.npu.ci

@@ -1,13 +1,11 @@
-ARG VLLM_ASCEND_IMAGE=quay.io/atlas-ci/vllm-ascend
-ARG VLLM_ASCEND_TAG=v0.26.0
+ARG VLLM_ASCEND_IMAGE=quay.io/ascend/vllm-ascend
+ARG VLLM_ASCEND_TAG=nightly-main
 FROM ${VLLM_ASCEND_IMAGE}:${VLLM_ASCEND_TAG}
 
 ARG APP_DIR=/vllm-workspace/vllm-omni
 WORKDIR ${APP_DIR}
 COPY . .
 
-RUN sed -i "s@^\(deb.*\)https\?://[a-z0-9.-]*\.ubuntu\.com@\1http://cache-service.nginx-pypi-cache.svc.cluster.local:8081@g" /etc/apt/sources.list
-
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
         ffmpeg \
@@ -35,16 +33,8 @@ RUN source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
       "psutil>=7.2.0" \
       "soundfile>=0.13.1" \
       "pyttsx3>=2.99" \
-      "opencc==1.4.1" \
       "opencv-python>=4.12.0.88" \
       "imageio[ffmpeg]>=0.6.0" \
-      "jiwer>=3.0.0" \
-      "zhon>=2.0.0" \
-      "zhconv>=1.4.2" \
-      "funasr>=1.0.0" \
-      "hyperpyyaml>=1.2.0" \
-      "onnx>=1.16.0" \
-      "librosa>=0.10.0" \
       --index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple \
       --trusted-host cache-service.nginx-pypi-cache.svc.cluster.local \
       --default-timeout=60 && \
@@ -53,17 +43,6 @@ RUN source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
       --trusted-host cache-service.nginx-pypi-cache.svc.cluster.local \
       --default-timeout=60
 
-# MiniCPM-o 4.5 Stage-2 vocoder assets: s3tokenizer plus flashcosyvoice, which
-# is shipped by step-audio2. --no-deps keeps pip from replacing the Ascend torch
-# build and from shadowing onnxruntime-cann with the vanilla onnxruntime wheel;
-# the deps these two actually need are installed above.
-RUN python3 -m pip install -v --no-deps \
-      "s3tokenizer>=0.3.0" \
-      "step-audio2>=1.0.0" \
-      --index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple \
-      --trusted-host cache-service.nginx-pypi-cache.svc.cluster.local \
-      --default-timeout=60
-
 ENV VLLM_WORKER_MULTIPROC_METHOD=spawn
 
 ENTRYPOINT []

docker/Dockerfile.npu.ci.a3

@@ -1,13 +1,11 @@
-ARG VLLM_ASCEND_IMAGE=quay.io/atlas-ci/vllm-ascend
-ARG VLLM_ASCEND_TAG=v0.26.0-a3
+ARG VLLM_ASCEND_IMAGE=quay.io/ascend/vllm-ascend
+ARG VLLM_ASCEND_TAG=nightly-main-a3
 FROM ${VLLM_ASCEND_IMAGE}:${VLLM_ASCEND_TAG}
 
 ARG APP_DIR=/vllm-workspace/vllm-omni
 WORKDIR ${APP_DIR}
 COPY . .
 
-RUN sed -i "s@^\(deb.*\)https\?://[a-z0-9.-]*\.ubuntu\.com@\1http://cache-service.nginx-pypi-cache.svc.cluster.local:8081@g" /etc/apt/sources.list
-
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
         ffmpeg \
@@ -35,17 +33,8 @@ RUN source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
       "psutil>=7.2.0" \
       "soundfile>=0.13.1" \
       "pyttsx3>=2.99" \
-      "opencc==1.4.1" \
       "opencv-python>=4.12.0.88" \
       "imageio[ffmpeg]>=0.6.0" \
-      "jiwer>=3.0.0" \
-      "zhon>=2.0.0" \
-      "zhconv>=1.4.2" \
-      "funasr>=1.0.0" \
-      "hyperpyyaml>=1.2.0" \
-      "onnx>=1.16.0" \
-      "librosa>=0.10.0" \
-      "torchmetrics>=1.9.0" \
       --index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple \
       --trusted-host cache-service.nginx-pypi-cache.svc.cluster.local \
       --default-timeout=60 && \
@@ -54,17 +43,6 @@ RUN source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
       --trusted-host cache-service.nginx-pypi-cache.svc.cluster.local \
       --default-timeout=60
 
-# MiniCPM-o 4.5 Stage-2 vocoder assets: s3tokenizer plus flashcosyvoice, which
-# is shipped by step-audio2. --no-deps keeps pip from replacing the Ascend torch
-# build and from shadowing onnxruntime-cann with the vanilla onnxruntime wheel;
-# the deps these two actually need are installed above.
-RUN python3 -m pip install -v --no-deps \
-      "s3tokenizer>=0.3.0" \
-      "step-audio2>=1.0.0" \
-      --index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple \
-      --trusted-host cache-service.nginx-pypi-cache.svc.cluster.local \
-      --default-timeout=60
-
 ENV VLLM_WORKER_MULTIPROC_METHOD=spawn
 
 ENTRYPOINT []

docker/Dockerfile.rocm

@@ -1,4 +1,4 @@
-ARG BASE_IMAGE=vllm/vllm-openai-rocm:v0.26.0
+ARG BASE_IMAGE=vllm/vllm-openai-rocm:v0.23.0
 FROM ${BASE_IMAGE} AS base
 
 # Declare a variable to know if we want to use the nightly build or the stable build.
@@ -10,14 +10,14 @@ FROM ${BASE_IMAGE} AS base
 #    we should swap over to use stable release ASAP.
 #    We should avoid relying on custom commits.
 ARG USE_NIGHTLY_BUILD=0
-ARG VLLM_VERSION_OR_COMMIT_HASH=v0.26.0
+ARG VLLM_VERSION_OR_COMMIT_HASH=89138b21cc246ae944c741d5c399c148e2b770ab
 ARG ARG_PYTORCH_ROCM_ARCH
 ENV PYTORCH_ROCM_ARCH=${ARG_PYTORCH_ROCM_ARCH:-${PYTORCH_ROCM_ARCH}}
 
 ARG COMMON_WORKDIR=/app
 WORKDIR ${COMMON_WORKDIR}
 
-# Step 1: Setup - Install system dependencies.
+# Step 1: Setup - Install system dependencies
 # Need to include ffmpeg because vllm rocm upstream docker image
 # does not include it.
 RUN apt-get update && \
@@ -38,26 +38,6 @@ RUN if [ "${USE_NIGHTLY_BUILD}" = "1" ]; then \
         rm -rf vllm/.git; \
     fi
 
-# Build TorchCodec after any optional nightly vLLM reinstall so it targets the
-# final torch/ROCm ABI. Keep this layer before COPY . so source changes do not
-# invalidate the expensive native build. FFmpeg development headers are
-# removed after the wheel is built.
-COPY tools/install_torchcodec_rocm.sh /tmp/install_torchcodec_rocm.sh
-RUN --mount=type=cache,target=/root/.cache/pip \
-    --mount=type=cache,target=/root/.cache/torchcodec-wheels \
-    apt-get update && \
-    apt-get install -y --no-install-recommends \
-        pkg-config \
-        libavcodec-dev libavformat-dev libavutil-dev \
-        libswscale-dev libavdevice-dev libavfilter-dev libswresample-dev && \
-    bash /tmp/install_torchcodec_rocm.sh && \
-    apt-get purge -y \
-        libavcodec-dev libavformat-dev libavutil-dev \
-        libswscale-dev libavdevice-dev libavfilter-dev libswresample-dev && \
-    rm -f /tmp/install_torchcodec_rocm.sh && \
-    apt-get clean && \
-    rm -rf /var/lib/apt/lists/*
-
 # Step 3: Copy vllm-omni code and install without uv
 RUN mkdir -p ${COMMON_WORKDIR}/vllm-omni
 COPY . ${COMMON_WORKDIR}/vllm-omni
@@ -79,10 +59,7 @@ RUN printf '%s\n' \
     '    os._exit(_exit_code)' \
     > ${COMMON_WORKDIR}/vllm-omni/conftest.py
 
-RUN cd ${COMMON_WORKDIR}/vllm-omni && \
-    uv pip install --python "$(python3 -c 'import sys; print(sys.executable)')" \
-        --no-cache-dir ".[dev]" --no-build-isolation && \
-    python3 -c "from torchcodec.decoders import VideoDecoder"
+RUN cd ${COMMON_WORKDIR}/vllm-omni && uv pip install --python "$(python3 -c 'import sys; print(sys.executable)')" --no-cache-dir ".[dev]" --no-build-isolation
 
 RUN ln -sf /usr/bin/python3 /usr/bin/python
 

docker/Dockerfile.xpu

@@ -88,7 +88,7 @@ ENV UV_EXTRA_INDEX_URL=${PIP_EXTRA_INDEX_URL}
 ENV UV_INDEX_STRATEGY="unsafe-best-match"
 ENV UV_LINK_MODE="copy"
 
-ARG VLLM_VERSION=v0.26.0
+ARG VLLM_VERSION=v0.23.0
 RUN git clone -b ${VLLM_VERSION} https://github.com/vllm-project/vllm
 WORKDIR /workspace/vllm
 
@@ -101,15 +101,13 @@ RUN --mount=type=cache,target=/root/.cache/uv \
     source /opt/intel/oneapi/ccl/2021.15/env/vars.sh --force && \
     export CMAKE_PREFIX_PATH="$(python3 -c 'import site; print(site.getsitepackages()[0])'):${CMAKE_PREFIX_PATH}"
 
-ENV LD_LIBRARY_PATH="/opt/intel/oneapi/ccl/2021.15/lib:$LD_LIBRARY_PATH:/usr/local/lib/"
+ENV LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/lib/"
 
 ENV VLLM_TARGET_DEVICE=xpu
 ENV VLLM_WORKER_MULTIPROC_METHOD=spawn
 
 RUN --mount=type=cache,target=/root/.cache/uv \
-    uv pip uninstall triton triton-xpu && \
-    uv pip install triton-xpu==3.7.1 && \
-    uv pip install --no-build-isolation --no-deps .
+    uv pip install --no-build-isolation .
 
 CMD ["/bin/bash"]
 
@@ -145,11 +143,12 @@ COPY . .
 
 ENV VLLM_OMNI_TARGET_DEVICE=xpu
 RUN --mount=type=cache,target=/root/.cache/uv \
+    echo "triton @ https://example.com/blocked" > /tmp/vllm_omni_constraints.txt && \
+    PIP_CONSTRAINT=/tmp/vllm_omni_constraints.txt \
     uv pip install --no-build-isolation \
     --extra-index-url https://mirrors.ustc.edu.cn/pypi/simple/ \
     ".[dev]" && \
-    uv pip uninstall triton oneccl oneccl-devel && \
-    uv pip install triton-xpu==3.7.1 --reinstall
+    uv pip uninstall oneccl oneccl-devel
 
 FROM vllm-omni AS vllm-omni-openai
 

docs/.nav.yml

@@ -16,12 +16,15 @@ nav:
   - Examples:
     - examples/README.md
     - Offline Inference:
+      - BAGEL-7B-MoT: user_guide/examples/offline_inference/bagel.md
       - GLM-Image Multistage End-to-End Inference: user_guide/examples/offline_inference/glm_image.md
+      - Helios Video Generation: user_guide/examples/offline_inference/helios.md
       - HunyuanImage-3.0 Image-to-Text Inference: user_guide/examples/offline_inference/hunyuan_image3.md
       - Image-To-Image: user_guide/examples/offline_inference/image_to_image.md
       - Image-To-Video: user_guide/examples/offline_inference/image_to_video.md
       - InternVLA-A1: user_guide/examples/offline_inference/internvla_a1.md
-      - MiMo-Audio: user_guide/examples/offline_inference/mimo_audio.md
+      - MammothModa2-Preview: user_guide/examples/offline_inference/mammothmodal2_preview.md
+      - MiMo-Audio Offline Inference: user_guide/examples/offline_inference/mimo_audio.md
       - Qwen2.5-Omni: user_guide/examples/offline_inference/qwen2_5_omni.md
       - Qwen3-Omni: user_guide/examples/offline_inference/qwen3_omni.md
       - Text-To-Audio: user_guide/examples/offline_inference/text_to_audio.md
@@ -36,7 +39,7 @@ nav:
       - GLM-Image Online Serving: user_guide/examples/online_serving/glm_image.md
       - Image-To-Image: user_guide/examples/online_serving/image_to_image.md
       - Image-To-Video: user_guide/examples/online_serving/image_to_video.md
-      - MiMo-Audio: user_guide/examples/online_serving/mimo_audio.md
+      - Online serving Example of vLLM-Omni for MiMo-Audio: user_guide/examples/online_serving/mimo_audio.md
       - Qwen2.5-Omni: user_guide/examples/online_serving/qwen2_5_omni.md
       - Qwen3-Omni: user_guide/examples/online_serving/qwen3_omni.md
       - Text-To-Speech: user_guide/examples/online_serving/text_to_speech.md
@@ -54,7 +57,6 @@ nav:
     - Diffusion Features:
       - Overview: user_guide/diffusion_features.md
       - Feature Compatibility: user_guide/feature_compatibility.md
-      - Regional Compilation: user_guide/diffusion/regional_compilation.md
       - Cache Acceleration:
         - TeaCache: user_guide/diffusion/cache_acceleration/teacache.md
         - Cache-DiT: user_guide/diffusion/cache_acceleration/cache_dit.md
@@ -67,18 +69,17 @@ nav:
         - Hybrid Sharded Data Parallel: user_guide/diffusion/parallelism/hsdp.md
         - Sequence Parallel: user_guide/diffusion/parallelism/sequence_parallel.md
         - Tensor Parallel: user_guide/diffusion/parallelism/tensor_parallel.md
-        - VAE Parallelism: user_guide/diffusion/parallelism/vae_parallelism.md
-      - CPU Offloading: user_guide/diffusion/cpu_offload.md
+        - VAE Patch Parallel: user_guide/diffusion/parallelism/vae_patch_parallel.md
+      - CPU Offloading: user_guide/diffusion/cpu_offload_diffusion.md
       - LoRA: user_guide/diffusion/lora.md
       - Custom Pipeline: features/custom_pipeline.md
-      - Execution Modes: user_guide/diffusion/execution_modes.md
+      - Step Execution: user_guide/diffusion/step_execution.md
     - Quantization:
-      - Quantization: user_guide/quantization/overview.md
+      - Overview: user_guide/quantization/overview.md
       - Online Quantization: user_guide/quantization/online.md
       - Quantized KV Cache: user_guide/quantization/quantized_kvcache.md
       - FP8 W8A8: user_guide/quantization/fp8.md
       - Int8 W8A8: user_guide/quantization/int8.md
-      - BitsAndBytes W4: user_guide/quantization/bitsandbytes.md
       - MXFP8 W8A8: user_guide/quantization/mxfp8.md
       - MXFP4 W4A4: user_guide/quantization/mxfp4.md
       - ModelOpt: user_guide/quantization/modelopt.md
@@ -101,49 +102,29 @@ nav:
   - CI: contributing/ci
   - Design Documents:
     - design/index.md
-    - Architecture Documents:
-      - Architecture Overview: design/architecture_overview.md
+    - design/architecture_overview.md
     - Feature Design:
-      - Runtime and Stage Execution:
-        - Disaggregated Inference: design/feature/disaggregated_inference.md
-        - Async Chunk: design/feature/async_chunk.md
-        - Async Diffusion Output: design/feature/async_diffusion_output.md
-        - Async Omni Output Materialization: design/feature/omni_async_output_materialization.md
-        - Automatic Prefix Caching: design/feature/prefix_caching.md
-      - Communication:
-        - OmniConnector Implementations:
-          - Mooncake Store Connector: design/feature/omni_connectors/mooncake_store_connector.md
-          - Mooncake Transfer Engine Connector: design/feature/omni_connectors/mooncake_transfer_engine_connector.md
-          - Mori Transfer Engine Connector: design/feature/omni_connectors/mori_transfer_engine_connector.md
-          - Shared Memory Connector: design/feature/omni_connectors/shared_memory_connector.md
-          - Yuanrong Store Connector: design/feature/omni_connectors/yuanrong_connector.md
-          - Yuanrong Transfer Engine Connector: design/feature/omni_connectors/yuanrong_transfer_engine_connector.md
-      - Diffusion Acceleration:
-        - Parallelism:
-          - CFG-Parallel: design/feature/cfg_parallel.md
-          - Expert Parallel: design/feature/expert_parallel.md
-          - Hybrid Sharded Data Parallel (HSDP): design/feature/hsdp.md
-          - Pipeline Parallel: design/feature/pipeline_parallel.md
-          - Sequence Parallel: design/feature/sequence_parallel.md
-          - Tensor Parallel: design/feature/tensor_parallel.md
-          - VAE Patch Parallelism: design/feature/vae_parallel.md
-        - Attention Backends:
-          - Skip-Softmax: design/feature/skip_softmax.md
-        - Quantization: design/feature/quantization.md
-        - Cache-DiT: design/feature/cache_dit.md
-        - TeaCache: design/feature/teacache.md
-        - Diffusion Continuous Batching: design/feature/diffusion_continuous_batching.md
-        - Distributed Layerwise Offload: design/feature/distributed_layerwise_offload.md
-    - Infrastructure and Performance:
-      - Prometheus Metrics: design/metrics.md
-      - Speech Generation Performance Optimizations: design/qwen3_omni_tts_performance_optimization.md
+      - design/feature/disaggregated_inference.md
+      - design/feature/ray_based_execution.md
+      - design/feature/omni_connectors/
+      - design/feature/prefix_caching.md
+      - design/feature/cfg_parallel.md
+      - design/feature/expert_parallel.md
+      - design/feature/sequence_parallel.md
+      - design/feature/tensor_parallel.md
+      - design/feature/vae_parallel.md
+      - design/feature/hsdp.md
+      - design/feature/cache_dit.md
+      - design/feature/teacache.md
+      - design/feature/async_chunk.md
+      - design/feature/vae_parallel.md
+      - design/feature/diffusion_step_execution.md
+      - design/feature/diffusion_continuous_batching.md
     - Module Design:
-      - Runtime Modules:
-        - AR Module: design/module/ar_module.md
-        - DIT Module: design/module/dit_module.md
-      - Orchestration:
-        - Entrypoint Module: design/module/entrypoint_module.md
-        - AsyncOmni Architecture: design/module/async_omni_architecture.md
+      - design/module/ar_module.md
+      - design/module/dit_module.md
+      - design/module/entrypoint_module.md
+      - design/module/async_omni_architecture.md
   - Docs Guide: contributing/DOCS_GUIDE.md
 - API Reference:
   - api/README.md

docs/README.md

@@ -28,9 +28,9 @@ Easy, fast, and cheap omni-modality model serving for everyone
 
 [vLLM](https://github.com/vllm-project/vllm) was originally designed to support large language models for text-based autoregressive generation tasks. vLLM-Omni is a framework that extends its support for omni-modality model inference and serving:
 
-- **Omni-modality**: Text, image, audio, video, and action data processing
+- **Omni-modality**: Text, image, video, and audio data processing
 - **Non-autoregressive Architectures**: extend the AR support of vLLM to Diffusion Transformers (DiT) and other parallel generation models
-- **Heterogeneous outputs**: from traditional text generation to multimodal and action outputs
+- **Heterogeneous outputs**: from traditional text generation to multimodal outputs
 
 <p align="center">
   <picture>
@@ -52,14 +52,11 @@ vLLM-Omni is flexible and easy to use with:
 - Tensor, pipeline, data and expert parallelism support for distributed inference
 - Streaming outputs
 - OpenAI-compatible API server
-- Full-duplex realtime serving with streaming audio input and output (experimental)
 
 vLLM-Omni seamlessly supports most popular open-source models on HuggingFace, including:
 
-- Omni-modality models (e.g. Qwen3-Omni, MiniCPM-o 4.5, Cosmos3, HunyuanImage, BAGEL)
-- TTS models (e.g. Qwen3-TTS, VoxCPM2, Ming-Omni-TTS, CosyVoice3)
-- Diffusion models — image, video, and audio generation (e.g. MiniMax H3, Qwen-Image, Wan2.2, FLUX)
-- Robot-policy and action models (e.g. GR00T-N1.7, DreamZero-DROID, InternVLA-A1, Cosmos3 action policy)
+- Omni-modality models (e.g. Qwen2.5-Omni, Qwen3-Omni)
+- Multi-modality generation models (e.g. Qwen-Image)
 
 For more information, checkout the following:
 

docs/api/README.md

@@ -46,22 +46,23 @@ Engine classes for offline and online inference.
 - [vllm_omni.engine.PromptEmbedsPayload][]
 - [vllm_omni.engine.arg_utils.OmniEngineArgs][]
 - [vllm_omni.engine.async_omni_engine.AsyncOmniEngine][]
+- [vllm_omni.engine.mm_outputs.MultimodalCompletionOutput][]
+- [vllm_omni.engine.mm_outputs.MultimodalPayload][]
 - [vllm_omni.engine.orchestrator.Orchestrator][]
 - [vllm_omni.engine.orchestrator.OrchestratorRequestState][]
+- [vllm_omni.engine.output_modality.OutputModality][]
+- [vllm_omni.engine.output_modality.TensorAccumulationStrategy][]
+- [vllm_omni.engine.output_processor.MultimodalOutputProcessor][]
+- [vllm_omni.engine.output_processor.OmniRequestState][]
 - [vllm_omni.engine.stage_engine_core_client.StageEngineCoreClient][]
 - [vllm_omni.engine.stage_init_utils.StageMetadata][]
 - [vllm_omni.engine.stage_init_utils.StartedLlmStage][]
-- [vllm_omni.outputs.mm_outputs.MultimodalCompletionOutput][]
-- [vllm_omni.outputs.mm_outputs.MultimodalPayload][]
-- [vllm_omni.outputs.output_modality.OutputModality][]
-- [vllm_omni.outputs.output_modality.TensorAccumulationStrategy][]
-- [vllm_omni.outputs.output_processor.MultimodalOutputProcessor][]
-- [vllm_omni.outputs.output_processor.OmniRequestState][]
 
 ## Core
 
 Core scheduling and caching components.
 
+- [vllm_omni.core.sched.omni_ar_scheduler.KVCacheTransferData][]
 - [vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler][]
 - [vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler][]
 - [vllm_omni.core.sched.output.OmniCachedRequestData][]
@@ -130,7 +131,7 @@ Worker classes and model runners for distributed inference.
 - [vllm_omni.diffusion.worker.diffusion_worker.DiffusionWorker][]
 - [vllm_omni.diffusion.worker.diffusion_worker.WorkerProc][]
 - [vllm_omni.diffusion.worker.diffusion_worker.WorkerWrapperBase][]
-- [vllm_omni.diffusion.worker.utils.StepRequestState][]
+- [vllm_omni.diffusion.worker.utils.DiffusionRequestState][]
 - [vllm_omni.diffusion.worker.utils.RunnerOutput][]
 - [vllm_omni.platforms.npu.worker.npu_ar_model_runner.ExecuteModelState][]
 - [vllm_omni.platforms.npu.worker.npu_ar_model_runner.NPUARModelRunner][]

docs/cli/README.md

@@ -18,12 +18,12 @@ Specify the port:
 vllm serve Qwen/Qwen2.5-Omni-7B --omni --port 8091
 ```
 
-Load a custom deploy configuration with `--deploy-config`:
-
+If you have custom stage configs file, launch the server with command below
 ```bash
-vllm serve Qwen/Qwen2.5-Omni-7B --omni --deploy-config /path/to/deploy_config.yaml
+vllm serve Qwen/Qwen2.5-Omni-7B --omni --stage-configs-path /path/to/stage_configs_file
 ```
 
+
 ## bench
 
 Run benchmark tests for online serving throughput.

docs/cli/bench/serve.md

@@ -2,7 +2,7 @@
 The vllm bench command launches the vLLM-Omni benchmark to evaluate the performance of multimodal models.
 
 ## Notes
-vLLM-Omni registers the `openai-chat-omni`, `openai-audio-speech`, `openai-image-edits-omni`, and `daily-omni` serving benchmark backends.
+We currently only support using the "openai-chat-omni" and "openai-image-edits-omni" backend.
 
 ## Basic Parameter Description
 You can use `vllm bench serve --omni --help=all` to get descriptions of all parameters. The commonly used parameters are described below:
@@ -10,7 +10,7 @@ You can use `vllm bench serve --omni --help=all` to get descriptions of all para
   Enable Omni (multimodal) mode, supporting multimodal inputs and outputs such as images, videos, and audio.
 
 - `--backend`
-  Specify the backend adapter. vLLM-Omni adds `openai-chat-omni`, `openai-audio-speech`, `openai-image-edits-omni`, and `daily-omni` to the upstream vLLM backend choices.
+  Specify the backend adapter as openai-chat-omni, using OpenAI Chat compatible API behavior as the protocol. Currently only openai-chat-omni is supported.
 
 - `--model`
   The model identifier to load, filled according to the models supported by vLLM-Omni.
@@ -55,7 +55,7 @@ You can use `vllm bench serve --omni --help=all` to get descriptions of all para
                     'Allowed metric names are "ttft", "tpot", "itl", "ttfc", "tpoc", "icl", '
                     '"tpop", "e2el", "audio_ttfp", "audio_rtf", "audio_duration". '
 
-- `--print-stage`
+- `-print-stage`
 Print per-stage benchmark metrics for --omni serving when stage metrics are returned by the server. Disabled by default.
 
 - `--save-result`
@@ -312,7 +312,7 @@ vllm bench serve \
   --ignore-eos \
   --percentile-metrics ttft,tpot,itl \
   --random-output-len 2 \
-  --extra-body '{"modalities": ["text"]}'
+  --extra_body '{"modalities": ["text"]}'
 ```
 
 If successful, you will see the following output:
@@ -390,7 +390,7 @@ vllm bench serve --omni \
   --ignore-eos \
   --percentile-metrics ttft,tpot,itl,e2el,audio_ttfp,audio_rtf,ttfc,tpoc,icl \
   --random-output-len 900 \
-  --extra-body '{"modalities": ["text","audio"]}' \
+  --extra_body '{"modalities": ["text","audio"]}' \
   --print-stage
 ```
 

docs/cli/serve.md

@@ -8,6 +8,8 @@ The stage-based CLI is designed for deployments that require launching each pipe
 - For **migrated models** that utilize the bundled deployment YAML configurations located in
   `vllm_omni/deploy/`, the `--deploy-config` flag is only required to override the default configuration. By default, executing `vllm serve MODEL --omni ...`
   automatically loads the bundled deployment configuration.
+- For **legacy models** utilizing configuration files located in
+  `vllm_omni/model_executor/stage_configs/`, the `--stage-configs-path` parameter remains mandatory.
 
 Example: Initializing Stage 0 (Orchestrator and API Server):
 The commands below show a common device mapping where Stage 0 uses GPU 0 and
@@ -31,7 +33,7 @@ CUDA_VISIBLE_DEVICES=1 vllm serve Qwen/Qwen3-Omni-30B-A3B-Instruct --omni \
     --omni-master-port 26000
 ```
 
-When utilizing a custom deployment YAML, append `--deploy-config /path/to/override.yaml` to each command execution.
+When utilizing a custom deployment YAML based on the new schema, append `--deploy-config /path/to/override.yaml` to each command execution. Conversely, for legacy models, substitute this parameter with `--stage-configs-path /path/to/stage_configs.yaml`.
 
 In the standard execution paradigm, the `--stage-overrides` argument is utilized to apply stage-specific configurations from a single CLI command.
 However, under the **stage-based CLI** paradigm, where each process strictly encapsulates a single stage, it is recommended to specify tuning parameters directly via discrete command-line flags for the respective stage, rather than constructing a composite `--stage-overrides` JSON string.

docs/community/governance.md

@@ -33,21 +33,16 @@ Lead maintainers are responsible for the overall direction and strategy of the p
 
 Committers have write access and merge rights. They typically have deep expertise in specific areas of this project and shepherd the community contributions:
 
-- [@alex-jw-brooks](https://github.com/alex-jw-brooks): Frontend, Configuration, and Entrypoint Support
-- [@amy-why-3459](https://github.com/amy-why-3459): Omni Modality Model and World Model Support
 - [@david6666666](https://github.com/david6666666): Quantization and Community Relationship
-- [@gcanlin](https://github.com/gcanlin): Hardware plugin, NPU integration, and TTS Model Support
+- [@gcanlin](https://github.com/gcanlin): Hardware plugin and NPU integration
 - [@Isotr0py](https://github.com/Isotr0py): Diffusion and Quantization
 - [@linyueqian](https://github.com/linyueqian): TTS and Omni Support
 - [@lishunyang12](https://github.com/lishunyang12): Quantization and Configuration
-- [@NickCao](https://github.com/NickCao): CI, Tests, Entrypoint, and TTS Model Support
 - [@princepride](https://github.com/princepride): Diffusion and Omni Support
 - [@RuixiangMa](https://github.com/RuixiangMa): Diffusion models, parallel, cache, and docs
 - [@SamitHuang](https://github.com/SamitHuang): RL, Diffusion, and cache
-- [@Sy0307](https://github.com/Sy0307): TTS Model and AR Engine Support
 - [@tzhouam](https://github.com/tzhouam): Engine and New Model Support
 - [@wtomin](https://github.com/wtomin): Diffusion models, parallel, and docs
-- [@xuechendi](https://github.com/xuechendi): Diffusion and Intel GPU Support
 - [@ZeldaHuang](https://github.com/ZeldaHuang): Omni Support
 - [@ZJY0516](https://github.com/ZJY0516): Diffusion attention backend, kernel fusion, and CustomOp
 - [@yuanheng-zhao](https://github.com/yuanheng-zhao): Diffusion cache, offload, and Omni Support
@@ -58,7 +53,7 @@ This section breaks down diffusion-related responsibilities. If you have PRs tou
 
 ### Runtime-related
 
-- Diffusion models: @RuixiangMa, @wtomin, @xuechendi
+- Diffusion models: @RuixiangMa, @wtomin
 
 ### Optimization
 
@@ -66,7 +61,6 @@ This section breaks down diffusion-related responsibilities. If you have PRs tou
 - Attention backend: @ZJY0516
 - Cache: @yuanheng-zhao, @RuixiangMa, @SamitHuang
 - Offload: @yuanheng-zhao
-- Intel GPU support: @xuechendi
 - Quantization: @lishunyang12, @david6666666
 - Kernel fusion / communication-computation: @ZJY0516
 

docs/configuration/README.md

@@ -2,7 +2,7 @@
 
 This section lists the most common options for running vLLM-Omni.
 
-For options within a vLLM Engine, please refer to the [vLLM 0.26 configuration guide](https://docs.vllm.ai/en/v0.26.0/configuration/index.html).
+For options within a vLLM Engine. Please refer to [vLLM Configuration](https://docs.vllm.ai/en/v0.16.0/configuration/index.html)
 
 Currently, the main options are maintained by stage configs for each model.
 

docs/configuration/composable_parallel.md

@@ -1,224 +0,0 @@
-# Composable parallel strategies
-
-In vLLM-Omni, the `composable_parallel` system layers a declarative *parallel strategy* on top of the existing stage configs. A strategy YAML maps each pipeline stage (by its `model_stage` name) to a stack of axis declarations — one per parallelism dimension — that the engine translates into per-stage parallel sizing (`tensor_parallel_size`, `data_parallel_size`, `pipeline_parallel_size`, expert-parallel toggle, replica count) and, when relevant, an orchestrator-wide load-balancer policy.
-
-Strategies are intentionally narrow: each one only writes the axes it declares, leaving every other engine arg untouched. This makes a strategy file a small, reusable overlay you can swap independently of the deploy YAML (see `configuration/stage_configs.md`).
-
-!!! note
-    Composable parallel is **opt-in**. Passing `--strategy-config <path/to/strategy.yaml>` overlays the derived sizing onto the registry-merged stages *before* any CLI overrides are applied. When `--strategy-config` is omitted, stages remain exactly as their deploy YAML produces them.
-
-## Top-level schema reference
-
-A strategy file has a single top-level mapping, `strategies`, keyed by the `model_stage` name of each stage you want to overlay. Each value is a list of axis declarations (see [StrategySpec fields](#strategyspec-fields) below).
-
-| Field | Type | Required | Default | Description |
-|-------|------|----------|---------|-------------|
-| `strategies` | `dict[str, list]` | required | — | Maps `model_stage` name (e.g. `thinker`, `talker`, `code2wav`, `dit`) to a list of axis declarations. The key MUST be the stage name, not the integer `stage_id`. See `vllm_omni/config/composable_parallel/strategy_loader.py`. |
-
-### StrategySpec fields
-
-Each entry in a stage's list becomes one `StrategySpec` (see `vllm_omni/config/composable_parallel/spec.py`). The loader fills in `routing` and `aggregation` from per-kind defaults, so the file stays compact.
-
-| Field | Type | Required | Default | Description |
-|-------|------|----------|---------|-------------|
-| `axis` | str | required | — | Mesh-axis kind. One of `tp`, `dp`, `pp`, `ep`, `stage_replica` (wired today) or `sp_ulysses`, `sp_ring`, `cfg`, `vae_pp`, `hsdp`, `stage_pp`, `cp` (reserved; declaring one raises `AxisTranslationError` at translate time). See `MeshAxisKind` in `vllm_omni/config/composable_parallel/spec.py`. |
-| `size` | int | required | — | Axis degree. Must be `> 0`. |
-| `routing` | str | optional | per-kind default | Stateless routing policy. Accepted only for `dp` / `stage_replica`; other kinds reject it. Values: `random`, `round_robin`, `least_queue`. (`hash` is recognised but currently raises `NotImplementedError`.) See `RouteByStage` in `vllm_omni/config/composable_parallel/routing.py`. |
-| `l1_owner` | str | optional | per-kind default | Overrides which layer owns request routing for this axis. `engine` for `tp`/`dp`/`pp`/`ep`, `delegated` for `stage_replica`. Recorded into `shard_extension`. |
-| `name` | str | optional | axis kind | Human-readable label for this spec, used in error messages. |
-
-Two additional `StrategySpec` slots — `layer_hook_specs` and `kernel_specs` — let advanced authors bind layer-walker hook IDs (`LayerHookSpec`) or kernel IDs (`KernelSpec`) onto an axis. These are not exposed by the YAML loader today and are intended for programmatic construction. See `vllm_omni/config/composable_parallel/spec.py` for the dataclasses.
-
-### CLI flags
-
-| Flag | Description |
-|------|-------------|
-| `--strategy-config PATH` | Path to a `strategy.yaml`. Only takes effect on registry-based deploy paths (`--deploy-config` or the bundled default deploy YAML). Its derived sizing is overlaid onto the merged stages *before* CLI overrides. |
-| `--omni-lb-policy POLICY` | Orchestrator-wide load-balancer policy. Strategy-derived from any `stage_replica` axis; an explicit CLI value wins and must match the derived one or `AsyncOmniEngine` raises `Conflicting load-balancer policy` at construction. |
-
-## Sub-schemas
-
-### `axis`
-
-The set of recognised axis kinds is the `MeshAxisKind` enum in `vllm_omni/config/composable_parallel/spec.py`:
-
-```yaml
-# Translatable today (see `_SUPPORTED_KINDS` in translator.py).
-axis: tp              # tensor parallel — sets tensor_parallel_size
-axis: dp              # engine data parallel — sets data_parallel_size (vLLM's intra-engine DP LB)
-axis: pp              # pipeline parallel — sets pipeline_parallel_size
-axis: ep              # dense expert parallel — sets enable_expert_parallel=True; size must equal tp*dp
-axis: stage_replica   # omni-coordinator-level replica fan-out — sets num_replicas + omni_lb_policy
-
-# Reserved (raise AxisTranslationError today; declared in the type system for forward compat).
-axis: sp_ulysses
-axis: sp_ring
-axis: cfg
-axis: vae_pp
-axis: hsdp
-axis: stage_pp
-axis: cp
-```
-
-### `routing`
-
-The full closed hierarchy lives in `vllm_omni/config/composable_parallel/routing.py` (`Broadcast`, `PartitionByHash`, `ShardSequence`, `ShardSpatial`, `RouteByExpert`, `PipelineMicrobatch`, `DuplicateWithCondTag`, `RouteByStage`). For YAML authors only `RouteByStage` is exposed as a `routing:` string, and only for `dp` / `stage_replica` axes (other axes have fixed routing — `tp`/`ep` broadcast, `pp` microbatch). Recognised policies:
-
-```yaml
-routing: random         # uniform-random across replicas/DP ranks
-routing: round_robin    # cycle through replicas/DP ranks (the default for dp/stage_replica)
-routing: least_queue    # send to the replica with the smallest backlog
-# routing: hash         # recognised but raises NotImplementedError; no key-stable balancer yet
-```
-
-### `aggregation`
-
-Aggregation is filled in automatically by the loader (`_default_aggregation` in `vllm_omni/config/composable_parallel/strategy_loader.py`) from the axis kind, so users do not author it:
-
-| Axis kind | Aggregation pattern |
-|-----------|---------------------|
-| `tp`, `ep` | `TakeRank` (all ranks agree; pick one) |
-| `dp`, ep (in some flows) | `Union` (disjoint replica results) |
-| `pp` | `StitchPipeline` |
-| `stage_replica` | `FanInByStage` |
-
-Full pattern hierarchy: `TakeRank`, `Union`, `GatherDim`, `AllGather`, `StitchSpatial`, `StitchPipeline`, `Combine`, `FanInByStage` (see `vllm_omni/config/composable_parallel/aggregation.py`).
-
-## Precedence
-
-From highest to lowest, when the same axis is touched by multiple layers:
-
-1. **CLI overrides** (`--tensor-parallel-size`, `--stage-overrides`, `--omni-lb-policy`, etc.). A CLI value always wins; a warning is emitted whenever it overrides a strategy-declared axis (see `_reconcile_strategy_with_cli` in `vllm_omni/config/config_factory.py`).
-2. **Strategy YAML** (`--strategy-config`). Derived sizing is written onto the merged stages with *conflict-on-explicit* semantics — if the deploy YAML already set a value to something different, application raises `StrategyApplyError`.
-3. **Deploy YAML** (`--deploy-config` or the bundled `vllm_omni/deploy/<model_type>.yaml`).
-4. **Parser defaults.**
-
-After CLI overrides land, the device-layout guard re-runs (`check_device_layout`, called from `_reconcile_strategy_with_cli`) against the *effective* `tp * dp * pp * num_replicas` and the resolved `devices` string, so a `devices` value passed through `--stage-overrides` or a `--tensor-parallel-size` value that conflicts with the strategy's world size cannot slip past the pre-spawn check.
-
-## Worked example
-
-### Example 1 — `strategy_tp2.yaml`
-
-Shard the `thinker` stage with TP=2 on top of the bundled Qwen2.5-Omni deploy YAML. Routing is fixed (broadcast) for TP, so no `routing:` field is needed.
-
-```yaml
-# examples/offline_inference/qwen2_5_omni/strategy_tp2.yaml
-strategies:
-  thinker:
-    - axis: tp
-      size: 2
-```
-
-Launch:
-
-```bash
-python3 end2end.py --query-type text --num-prompts 6 \
-    --strategy-config strategy_tp2.yaml \
-    --stage-overrides '{"0": {"devices": "0,1"}, "1": {"devices": "2"}, "2": {"devices": "2"}}'
-```
-
-Effective `OmniParallelConfig` per stage after the overlay (see `OmniParallelConfig` in `vllm_omni/config/composable_parallel/translator.py`):
-
-| Stage (`model_stage`) | `tensor_parallel_size` | `data_parallel_size` | `pipeline_parallel_size` | `stage_replica_size` | `omni_lb_policy` |
-|-----------------------|------------------------|----------------------|--------------------------|----------------------|------------------|
-| `thinker`             | **2** (from strategy)  | 1                    | 1                        | 1                    | —                |
-| `talker`              | 1                      | 1                    | 1                        | 1                    | —                |
-| `code2wav`            | 1                      | 1                    | 1                        | 1                    | —                |
-
-### Example 2 — `strategy_stage_replica.yaml`
-
-Replicate the `thinker` stage across two independent engines and let omni's StagePool round-robin requests across them.
-
-```yaml
-# examples/offline_inference/qwen2_5_omni/strategy_stage_replica.yaml
-strategies:
-  thinker:
-    - axis: stage_replica
-      size: 2
-      routing: round_robin
-```
-
-Launch:
-
-```bash
-python3 end2end.py --query-type text --num-prompts 6 \
-    --strategy-config strategy_stage_replica.yaml \
-    --stage-overrides '{"0": {"devices": "0,1"}, "1": {"devices": "2"}, "2": {"devices": "2"}}'
-```
-
-Effective per-stage config and the derived orchestrator knob:
-
-| Stage (`model_stage`) | `tensor_parallel_size` | `stage_replica_size` | `omni_lb_policy` (derived) |
-|-----------------------|------------------------|----------------------|----------------------------|
-| `thinker`             | 1                      | **2**                | **`round-robin`**          |
-| `talker`              | 1                      | 1                    | —                          |
-| `code2wav`            | 1                      | 1                    | —                          |
-
-The derived `omni_lb_policy = round-robin` is **not** written into any stage's engine args. Instead it is surfaced on `StrategyApplyResult.omni_lb_policy` and applied once at orchestrator construction by `AsyncOmniEngine._apply_strategy_lb_policy` (see `vllm_omni/engine/async_omni_engine.py`, lines 1107–1134). If you also pass `--omni-lb-policy round-robin` the values must match; passing a different value raises `Conflicting load-balancer policy`.
-
-## Per-field deep-dive
-
-### `name`
-
-Default: the axis kind (e.g. `"tp"`).
-
-Human-readable label that appears in `AxisTranslationError` and `StrategyApplyError` messages. Set it to disambiguate two stages declaring the same axis kind when scanning logs.
-
-### `mesh_axis` (`axis` + `size`)
-
-Default: no default — both fields are required.
-
-`axis` chooses the parallelism dimension; `size` is its degree. Five kinds are wired end-to-end today (`tp`, `dp`, `pp`, `ep`, `stage_replica`); the remaining kinds in `MeshAxisKind` are reserved and raise `AxisTranslationError` at translate time so a typo cannot silently become a no-op (see `_SUPPORTED_KINDS` in `vllm_omni/config/composable_parallel/translator.py`). For `ep`, the size MUST equal `tensor_parallel_size * data_parallel_size`.
-
-### `routing`
-
-Default: per-kind — `RouteByStage(round_robin)` for `dp` and `stage_replica`; `PipelineMicrobatch()` for `pp`; `Broadcast()` for `tp`/`ep` and any other kind.
-
-User-facing string only on `dp` and `stage_replica` axes. The four recognised values are `random`, `round_robin`, `least_queue`, `hash`; `hash` is rejected today with `NotImplementedError`. See `RouteByStage` in `vllm_omni/config/composable_parallel/routing.py`.
-
-### `l1_owner`
-
-Default: per-kind — `engine` for `tp`/`dp`/`pp`/`ep`, `delegated` for `stage_replica`.
-
-Overrides which layer owns request routing along this axis. `engine` means vLLM realizes routing intra-engine; `delegated` means the omni coordinator owns routing across independent replicas. Set this only when overriding the per-kind default in `_DEFAULT_L1_OWNER` (`vllm_omni/config/composable_parallel/translator.py`); invalid values fail fast in `_resolve_l1_owner`.
-
-### `stage_replica` and `omni_lb_policy`
-
-`stage_replica` is *not* a vLLM world dimension; it spins up `size` independent engine replicas of one stage, coordinated by the omni `StagePool` load balancer. The axis writes both:
-
-- The stage's `num_replicas` (a per-stage deploy knob).
-- The pipeline-wide `omni_lb_policy` string (an orchestrator-level knob).
-
-Because `omni_lb_policy` is read once at `AsyncOmniEngine` construction — not from per-stage engine args — only one value is allowed across the whole pipeline. If two stages declare conflicting policies, application raises (see `apply_strategy_specs` in `vllm_omni/config/composable_parallel/apply.py`, lines 253–260). To set this knob manually instead, use `--omni-lb-policy` on the CLI; see [the model stage docs](stage_configs.md) for how stage replicas plug into the orchestrator at runtime.
-
-### `model_stage` keys
-
-Default: no default — required for every entry under `strategies:`.
-
-Strategy keys MUST be `model_stage` *names* (strings), the same labels stage configs use for `engine_args.model_stage` (see [`model_stage` in `stage_configs.md`](stage_configs.md#engine_argsmodel_stage)). Integer `stage_id` values are not accepted; the resolver matches on `model_stage` exactly (`_resolve_stage` in `vllm_omni/config/composable_parallel/apply.py`).
-
-## Common pitfalls
-
-!!! warning "Using `stage_id` integers as keys"
-    Strategy keys are `model_stage` names (str), not integer stage IDs. A file that uses an integer index will raise at apply time:
-
-    ```text
-    StrategyApplyError: strategy role model_stage=0 did not match any stage;
-    available stages: 'thinker'(id=0), 'talker'(id=1), 'code2wav'(id=2)
-    ```
-
-    Use the `model_stage` name (e.g. `thinker`) instead.
-
-!!! warning "Mixing CLI overrides with a strategy on the same axis"
-    A strategy is meant to be the single writer for the axes it declares. If a CLI override (`--tensor-parallel-size 2`, `--stage-overrides '{"0": {"tensor_parallel_size": 2}}'`) sets the same axis, the **CLI value wins** and a warning is emitted from `_reconcile_strategy_with_cli`. The device-layout guard then re-runs against the effective layout, so a CLI override that breaks `tp * dp * pp * num_replicas` will still fail fast at config time.
-
-!!! warning "Conflicting `omni_lb_policy` across stages"
-    `omni_lb_policy` is a single pipeline-wide knob. If two `stage_replica` axes in different stages derive different policies, `apply_strategy_specs` raises:
-
-    ```text
-    StrategyApplyError: role 'talker' derives omni_lb_policy='least-queue-length'
-    but role 'thinker' already derived 'round-robin'. The load-balancer policy is
-    pipeline-wide; only one value is allowed.
-    ```
-
-    Make the policies match, drop one, or supply the value explicitly via `--omni-lb-policy`.

docs/configuration/pd_disaggregation.md

@@ -123,7 +123,7 @@ After inserting the extra thinker stage, renumber the remaining stages:
     runtime:
       devices: "2"
     engine_input_source: [1]
-    sync_process_input_func: vllm_omni.model_executor.stage_input_processors.qwen3_omni.thinker2talker_token_only
+    custom_process_input_func: vllm_omni.model_executor.stage_input_processors.qwen3_omni.thinker2talker
 
   - stage_id: 3
     runtime:
@@ -131,7 +131,7 @@ After inserting the extra thinker stage, renumber the remaining stages:
     engine_args:
       max_num_seqs: 1
     engine_input_source: [2]
-    sync_process_input_func: vllm_omni.model_executor.stage_input_processors.qwen3_omni.talker2code2wav_token_only
+    custom_process_input_func: vllm_omni.model_executor.stage_input_processors.qwen3_omni.talker2code2wav
 ```
 
 Compared with the default Qwen3-Omni config:
@@ -158,7 +158,7 @@ runtime:
 
 ```bash
 vllm serve Qwen/Qwen3-Omni-30B-A3B-Instruct --omni --port 8091 \
-    --deploy-config /path/to/qwen3_omni_pd.yaml
+    --stage-configs-path /path/to/qwen3_omni_pd.yaml
 ```
 
 ## Operational Notes