[Dependency] Upgrade Cache-DiT to 1.5.0

Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
2026-08-11 23:21:39 +08:00
428 changed files with 3789 additions and 35316 deletions
@@ -139,12 +139,7 @@ steps:
        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'
-        # h100_2, not h100_1: advanced_model asserts transcribe every audio
-        # response with Whisper, and tests/helpers/media.py only uses an
-        # accelerator for Whisper when a spare GPU exists. On a single card it
-        # grades on CPU: ~8-10 min per clip vs ~30 s, which alone overruns the
-        # 50-minute job timeout (build 2956).
-        mirror_hardwares: h100_2
+        mirror_hardwares: h100_1

      - label: "Omni · MiniCPM-o 4.5 Duplex Test"
        source_file_dependencies:
@@ -456,7 +456,7 @@ steps:
        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_ltx.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"
@@ -477,16 +477,10 @@ steps:
        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 · LTX Accuracy Test"
-        timeout_in_minutes: 180
-        commands:
-          - pytest -s -v 'tests/e2e/accuracy/ltx/test_ltx_official_similarity.py::test_ltx_matches_official[ltx2_3_i2v]' -m full_model --run-level full_model
-          - pytest -s -v 'tests/e2e/accuracy/ltx/test_ltx_official_similarity.py::test_ltx_matches_official[ltx23_two_stage_layer_fused_i2v]' -m full_model --run-level full_model
-        mirror_hardwares: h100_1
-
      - label: ":full_moon: Diffusion X2V · Perf Test"
        key: nightly-diffusion-x2v-performance
        timeout_in_minutes: 180
@@ -128,21 +128,6 @@ steps:
              image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
              always-pull: true
              propagate-environment: true
-              # vLLM 0.27 added a pre-flight capacity check in
-              # distributed/device_communicators/shm_broadcast.py. Without this
-              # the container gets Docker's 64 MiB default and every test in
-              # this step dies at engine start with
-              #   RuntimeError: Insufficient space in /dev/shm:
-              #                 240 MiB required, 64 MiB available
-              # (9 failed in 8.7s on build #2949). vLLM 0.26 had no such check
-              # and allocated regardless, so this step passed on main while
-              # being under-provisioned the whole time — 0.27 turned a silent
-              # risk into an explicit error.
-              #
-              # This is the only inline docker plugin block in .buildkite/cuda/;
-              # every other step inherits common/ci_mirror_hardwares.yml, which
-              # already sets 8gb. Matching that rather than inventing a value.
-              shm-size: "8gb"
              environment:
                - "HF_HOME=/fsx/hf_cache"
              volumes:
@@ -227,18 +212,6 @@ steps:
            "
        mirror_hardwares: h100_1

-      - label: "Omni · MiniCPM-o 4.5 Perf Test"
-        timeout_in_minutes: 60
-        source_file_dependencies:
-          - tests/dfx/perf/tests/test_minicpmo_4_5_ready.json
-          - vllm_omni/deploy/minicpmo_4_5.yaml
-          - vllm_omni/model_executor/models/minicpmo_4_5/
-          - vllm_omni/model_executor/stage_input_processors/minicpmo_4_5_omni.py
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_minicpmo_4_5_ready.json -m 'core_model and cuda'
-        mirror_hardwares: h100_1
-
      - label: "TTS · VoxCPM2 Test"
        source_file_dependencies:
          - tests/e2e/online_serving/test_voxcpm2_tts.py
@@ -10,7 +10,7 @@ steps:
          DOCKER_BUILDKIT: "1"
          # Buildkite will automatically replace this with the actual commit hash
          VLLM_IMAGE_TAG: "${BUILDKITE_COMMIT}"
-          VLLM_VERSION: "v0.27.0"
+          VLLM_VERSION: "v0.26.0"
        priority: 100
        timeout_in_minutes: 60
        soft_fail: false
@@ -34,23 +34,8 @@ docker_build() {
    docker build "${out[@]}" "$@" -f docker/Dockerfile.xpu .
 }

-build_vllm_base() (
-    local vllm_source_dir
-    vllm_source_dir=$(mktemp -d)
-    trap 'rm -rf "${vllm_source_dir}"' EXIT
-
-    git clone --depth 1 --branch "${VLLM_VERSION}" \
-        https://github.com/vllm-project/vllm "${vllm_source_dir}"
-
-    local out=("${export_args[@]/'{{IMAGE}}'/${base_image_name}}")
-    docker build "${out[@]}" \
-        --target vllm-openai \
-        -f "${vllm_source_dir}/docker/Dockerfile.xpu" \
-        "${vllm_source_dir}"
-)
-
 if [ -z "$(docker images -q "${base_image_name}")" ]; then
-    build_vllm_base
+    docker_build "${base_image_name}" --target vllm-base --build-arg "VLLM_VERSION=${VLLM_VERSION}"
 fi

 # Try building the docker image
@@ -17,15 +17,6 @@ steps:
          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
        commands:
          - pytest -s -v tests/e2e/online_serving/test_minicpmo_4_5_duplex.py -m 'core_model and npu and A3 and omni' --run-level 'core_model'
-      - label: "Omni · MiniCPM-o 4.5 Perf Test"
-        timeout_in_minutes: 60
-        mirror_hardwares: a3_npu_2
-        env:
-          VLLM_WORKER_MULTIPROC_METHOD: spawn
-          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
-        commands:
-          - export BENCHMARK_DIR=tests/dfx/perf/results
-          - pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_minicpmo_4_5_ready.json -m 'core_model and npu and A3 and omni'

  - group: ":card_index_dividers: TTS Model Test"
    key: ready-tts-group
@@ -40,11 +40,9 @@ steps:
    commands:
      - "bash .buildkite/release/scripts/publish-release-wheels-pypi.sh"

-  # Auto-skip when NIGHTLY=1 (scheduled); otherwise require manual unblock.
  - block: "Unblock to build release Docker images"
    depends_on: ~
    key: block-build-release-images
-    if: build.env("NIGHTLY") != "1"

  - group: "Build release Docker images"
    key: "build-release-images"
@@ -65,7 +63,7 @@ steps:
              --progress plain \
              -f docker/Dockerfile.cuda .
          - "docker push public.ecr.aws/q9t5s3a7/vllm-omni-release-repo:$BUILDKITE_COMMIT-$(uname -m)"
-          # Fallback default tag (no arch suffix) if arm64 build fails.
+          # re-tag to default image tag and push, just in case arm64 build fails
          - "docker tag public.ecr.aws/q9t5s3a7/vllm-omni-release-repo:$BUILDKITE_COMMIT-$(uname -m) public.ecr.aws/q9t5s3a7/vllm-omni-release-repo:$BUILDKITE_COMMIT"
          - "docker push public.ecr.aws/q9t5s3a7/vllm-omni-release-repo:$BUILDKITE_COMMIT"
        env:
@@ -88,26 +86,6 @@ steps:
        env:
          DOCKER_BUILDKIT: "1"

-  # NIGHTLY=1 only: promote ECR images to DockerHub nightly tags, then prune.
-  - label: "Publish nightly multi-arch image to DockerHub"
-    depends_on:
-      - build-release-image-x86
-      - build-release-image-arm64
-    if: build.env("NIGHTLY") == "1"
-    key: publish-nightly-images-dockerhub
-    agents:
-      queue: small_cpu_queue_release
-    commands:
-      - "bash .buildkite/release/scripts/publish-release-images.sh nightly"
-      - "bash .buildkite/release/scripts/cleanup-nightly-builds.sh"
-    plugins:
-      - docker-login#v3.0.0:
-          username: vllmbot
-          password-env: DOCKERHUB_TOKEN
-    env:
-      DOCKER_BUILDKIT: "1"
-      DOCKERHUB_USERNAME: "vllmbot"
-
  - block: "Publish release images to DockerHub"
    key: block-publish-release-images
    depends_on:
@@ -1,147 +0,0 @@
-#!/bin/bash
-# SPDX-License-Identifier: Apache-2.0
-#
-# Clean up old nightly builds from DockerHub (vllm/vllm-omni), keeping the
-# newest 14 tags with the nightly- prefix.
-
-set -ex
-
-TAG_PREFIX="nightly-"
-REPO="vllm/vllm-omni"
-REPO_API_URL="https://hub.docker.com/v2/repositories/${REPO}/tags"
-
-echo "Cleaning up tags with prefix: $TAG_PREFIX in repository: $REPO"
-
-if [ -z "$DOCKERHUB_TOKEN" ]; then
-  echo "Error: DOCKERHUB_TOKEN environment variable is not set"
-  exit 1
-fi
-
-if [ -z "$DOCKERHUB_USERNAME" ]; then
-  echo "Error: DOCKERHUB_USERNAME environment variable is not set"
-  exit 1
-fi
-
-echo "Getting DockerHub bearer token..."
-set +x
-BEARER_TOKEN=$(curl -s -X POST \
-  -H "Content-Type: application/json" \
-  -d "{\"username\": \"$DOCKERHUB_USERNAME\", \"password\": \"$DOCKERHUB_TOKEN\"}" \
-  "https://hub.docker.com/v2/users/login" | jq -r '.token')
-set -x
-
-if [ -z "$BEARER_TOKEN" ] || [ "$BEARER_TOKEN" = "null" ]; then
-  echo "Error: Failed to get DockerHub bearer token"
-  exit 1
-fi
-
-get_all_tags() {
-  local page=1
-  local all_tags=""
-  # Hub `name` is a partial match (not a strict prefix). Use it to narrow the
-  # list, then enforce startswith() client-side. Exhaust API pages; do not
-  # stop merely because a page has no matching tags after filtering.
-  local name_query
-  name_query=$(printf '%s' "$TAG_PREFIX" | jq -sRr @uri)
-
-  while true; do
-    set +x
-    local response
-    response=$(curl -s -H "Authorization: Bearer $BEARER_TOKEN" \
-      "$REPO_API_URL?page=$page&page_size=100&name=${name_query}")
-    set -x
-
-    local page_count
-    page_count=$(echo "$response" | jq -r '.results | length')
-    if [ -z "$page_count" ] || [ "$page_count" = "0" ] || [ "$page_count" = "null" ]; then
-      break
-    fi
-
-    local tags
-    tags=$(echo "$response" | jq -r --arg prefix "$TAG_PREFIX" \
-      '.results[] | select(.name | startswith($prefix)) | "\(.last_updated)|\(.name)"')
-
-    if [ -n "$tags" ]; then
-      all_tags="$all_tags$tags"$'\n'
-    fi
-    page=$((page + 1))
-  done
-
-  echo "$all_tags" | sort -r | cut -d'|' -f2
-}
-
-delete_tag() {
-  local tag_name="$1"
-  echo "Deleting tag: $tag_name"
-
-  local delete_url="https://hub.docker.com/v2/repositories/${REPO}/tags/$tag_name"
-  set +x
-  local response http_code body
-  response=$(curl -s -w "\n%{http_code}" -X DELETE \
-    -H "Authorization: Bearer $BEARER_TOKEN" "$delete_url")
-  set -x
-  http_code=$(printf '%s\n' "$response" | tail -n1)
-  body=$(printf '%s\n' "$response" | sed '$d')
-
-  if [ "$http_code" -ge 200 ] && [ "$http_code" -lt 300 ]; then
-    echo "Successfully deleted tag: $tag_name (HTTP ${http_code})"
-    return 0
-  fi
-
-  local detail=""
-  if echo "$body" | jq -e '.detail' > /dev/null 2>&1; then
-    detail=$(echo "$body" | jq -r '.detail')
-  elif [ -n "$body" ]; then
-    detail="$body"
-  else
-    detail="(empty response body)"
-  fi
-  echo "Error: Failed to delete tag $tag_name (HTTP ${http_code}): ${detail}"
-  return 1
-}
-
-echo "Fetching all tags from DockerHub..."
-all_tags=$(get_all_tags)
-
-if [ -z "$all_tags" ]; then
-  echo "No tags found to clean up"
-  exit 0
-fi
-
-total_tags=$(echo "$all_tags" | wc -l)
-echo "Found $total_tags tags"
-
-tags_to_keep=14
-tags_to_delete=$((total_tags - tags_to_keep))
-
-if [ $tags_to_delete -le 0 ]; then
-  echo "No tags need to be deleted (only $total_tags tags found, keeping $tags_to_keep)"
-  exit 0
-fi
-
-echo "Will delete $tags_to_delete old tags, keeping the newest $tags_to_keep"
-
-tags_to_delete_list=$(echo "$all_tags" | tail -n +$((tags_to_keep + 1)))
-
-if [ -z "$tags_to_delete_list" ]; then
-  echo "No tags to delete"
-  exit 0
-fi
-
-echo "Deleting old tags..."
-delete_failures=0
-while IFS= read -r tag; do
-  if [ -n "$tag" ]; then
-    if ! delete_tag "$tag"; then
-      delete_failures=$((delete_failures + 1))
-    fi
-    sleep 1
-  fi
-done <<< "$tags_to_delete_list"
-
-if [ "$delete_failures" -gt 0 ]; then
-  echo "Cleanup finished with ${delete_failures} deletion failure(s)"
-  exit 1
-fi
-
-echo "Cleanup completed successfully"
@@ -1,87 +1,50 @@
 #!/bin/bash
 # SPDX-License-Identifier: Apache-2.0
 #
-# Publish Docker images from ECR staging to DockerHub.
-#
-# Usage:
-#   publish-release-images.sh            # release mode (default)
-#   publish-release-images.sh release
-#   publish-release-images.sh nightly
-#
-# Release tags: latest, v<version>
-# Nightly tags: nightly, nightly-<commit>
+# Publish release Docker images from ECR to DockerHub.
+# Pulls per-arch images, tags with `latest` and versioned tags, pushes them,
+# then creates and pushes a multi-arch manifest.

 set -euo pipefail

-MODE="${1:-release}"
+RELEASE_VERSION=$(buildkite-agent meta-data get release-version --default "" | sed 's/^v//')
+if [ -z "${RELEASE_VERSION}" ]; then
+  echo "ERROR: release-version metadata not set"
+  exit 1
+fi

+COMMIT="$BUILDKITE_COMMIT"
 DOCKERHUB_REPO="vllm/vllm-omni"
 ECR_REPO="public.ecr.aws/q9t5s3a7/vllm-omni-release-repo"
-COMMIT="$BUILDKITE_COMMIT"
-PRIMARY_TAGS=()
-EXTRA_MANIFEST_TAGS=()
-
-case "${MODE}" in
-  release)
-    RELEASE_VERSION=$(buildkite-agent meta-data get release-version --default "" | sed 's/^v//')
-    if [ -z "${RELEASE_VERSION}" ]; then
-      echo "ERROR: release-version metadata not set"
-      exit 1
-    fi
-    PRIMARY_TAGS=("latest" "v${RELEASE_VERSION}")
-    ;;
-  nightly)
-    # Commit-pinned manifest reuses nightly-{x86_64,aarch64} arch images.
-    PRIMARY_TAGS=("nightly")
-    EXTRA_MANIFEST_TAGS=("nightly-${COMMIT}")
-    ;;
-  *)
-    echo "Usage: $0 {release|nightly}"
-    exit 2
-    ;;
-esac

 echo "========================================"
-echo "Publishing ${MODE} images to ${DOCKERHUB_REPO}"
+echo "Publishing release images v${RELEASE_VERSION}"
 echo "  Commit: ${COMMIT}"
-echo "  Primary tags: ${PRIMARY_TAGS[*]}"
-if [ ${#EXTRA_MANIFEST_TAGS[@]} -gt 0 ]; then
-  echo "  Extra manifests: ${EXTRA_MANIFEST_TAGS[*]}"
-fi
 echo "========================================"

+# Login to ECR to pull staging images
 aws ecr-public get-login-password --region us-east-1 | \
  docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7

-SRC_X86="${ECR_REPO}:${COMMIT}-x86_64"
-SRC_ARM="${ECR_REPO}:${COMMIT}-aarch64"
+docker pull "${ECR_REPO}:${COMMIT}-x86_64"
+docker pull "${ECR_REPO}:${COMMIT}-aarch64"

-docker pull "${SRC_X86}"
-docker pull "${SRC_ARM}"
+docker tag "${ECR_REPO}:${COMMIT}-x86_64" "${DOCKERHUB_REPO}:latest-x86_64"
+docker tag "${ECR_REPO}:${COMMIT}-x86_64" "${DOCKERHUB_REPO}:v${RELEASE_VERSION}-x86_64"
+docker push "${DOCKERHUB_REPO}:latest-x86_64"
+docker push "${DOCKERHUB_REPO}:v${RELEASE_VERSION}-x86_64"

-publish_manifest() {
-  local tag="$1"
-  local arch_tag_base="$2"
-  docker manifest rm "${DOCKERHUB_REPO}:${tag}" || true
-  docker manifest create \
-    "${DOCKERHUB_REPO}:${tag}" \
-    "${DOCKERHUB_REPO}:${arch_tag_base}-x86_64" \
-    "${DOCKERHUB_REPO}:${arch_tag_base}-aarch64"
-  docker manifest push "${DOCKERHUB_REPO}:${tag}"
-}
+docker tag "${ECR_REPO}:${COMMIT}-aarch64" "${DOCKERHUB_REPO}:latest-aarch64"
+docker tag "${ECR_REPO}:${COMMIT}-aarch64" "${DOCKERHUB_REPO}:v${RELEASE_VERSION}-aarch64"
+docker push "${DOCKERHUB_REPO}:latest-aarch64"
+docker push "${DOCKERHUB_REPO}:v${RELEASE_VERSION}-aarch64"

-for tag in "${PRIMARY_TAGS[@]}"; do
-  docker tag "${SRC_X86}" "${DOCKERHUB_REPO}:${tag}-x86_64"
-  docker tag "${SRC_ARM}" "${DOCKERHUB_REPO}:${tag}-aarch64"
-  docker push "${DOCKERHUB_REPO}:${tag}-x86_64"
-  docker push "${DOCKERHUB_REPO}:${tag}-aarch64"
-  publish_manifest "${tag}" "${tag}"
-done
-
-ARCH_BASE="${PRIMARY_TAGS[0]}"
-for tag in "${EXTRA_MANIFEST_TAGS[@]}"; do
-  publish_manifest "${tag}" "${ARCH_BASE}"
-done
+docker manifest rm "${DOCKERHUB_REPO}:latest" || true
+docker manifest rm "${DOCKERHUB_REPO}:v${RELEASE_VERSION}" || true
+docker manifest create "${DOCKERHUB_REPO}:latest" "${DOCKERHUB_REPO}:latest-x86_64" "${DOCKERHUB_REPO}:latest-aarch64"
+docker manifest create "${DOCKERHUB_REPO}:v${RELEASE_VERSION}" "${DOCKERHUB_REPO}:v${RELEASE_VERSION}-x86_64" "${DOCKERHUB_REPO}:v${RELEASE_VERSION}-aarch64"
+docker manifest push "${DOCKERHUB_REPO}:latest"
+docker manifest push "${DOCKERHUB_REPO}:v${RELEASE_VERSION}"

 echo ""
-echo "Successfully published ${MODE} images"
+echo "Successfully published release images for v${RELEASE_VERSION}"
@@ -1,6 +1,6 @@
 ---
 name: precheck-pr
-description: Self-check your branch before creating a PR — catch dead code, prevent new model-specific Python examples, verify accuracy/perf claims, validate PR title format, and confirm merge readiness. Use when the user says "precheck", "self review", "pre-submit check", or "check my PR before I open it." Never posts to GitHub.
+description: Self-check your branch before creating a PR — catch dead code, verify accuracy/perf claims, validate PR title format, and confirm merge readiness. Use when the user says "precheck", "self review", "pre-submit check", or "check my PR before I open it." Never posts to GitHub.
 ---

 # PR Pre-Check
@@ -65,8 +65,6 @@ Ask: "Quick mode or full mode?" Then walk the checklist for the detected PR type

 **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.

-**Also run the Examples-Policy check on every PR** using [references/examples-policy.md](references/examples-policy.md). Inspect only Python paths introduced by the diff. Treat a new model-, checkpoint-, vendor-, or family-specific Python example as blocking; do not report pre-existing example debt.
-
 ### Step 5: Print Report

 ```
@@ -79,7 +77,6 @@ Pre-check report for <branch>
  ─────────────────  ──────
  PR title format    ✓
  Code quality       ⚠ 1 broad except, 2 Any hints
-  Examples policy    ✗ new model-specific Python example
  PR desc integrity  ✓
  Registry/config    ✓
  Dead code          ⚠ 2 warnings
@@ -47,19 +47,6 @@ Diff-scoped sweep for five fragility patterns that are pervasive in the existing

 Roll the ⚠/✗ counts into the report as a single **Code quality** dimension row.

---
-
-## All PRs: Examples Policy (run on every PR, quick & full)
-
-Inspect Python files introduced under `examples/` using
-[examples-policy.md](examples-policy.md). This is a diff-scoped ratchet: existing
-model-specific examples may be edited or removed without triggering the check.
-
- [ ] **No new model-specific Python examples:** the diff adds, copies, or renames no Python file whose path or behavior is specific to one model, checkpoint, vendor, or model family. ✗ for any violation. Route inference through a shared task/protocol entrypoint, model request behavior through `vllm_omni/model_extras`, and commands through task docs or `recipes/`.
- [ ] **New generic example code is justified:** any new task-neutral Python file under `examples/` is an actual user-facing entrypoint rather than reusable production logic, a benchmark, an app, a setup tool, or a reproducer. ⚠ when the placement is questionable.
-
-Roll the result into the report as a separate **Examples policy** dimension.
-
 ### Conventions (eyeball — no grep)

 Judgment calls that don't grep cleanly — apply them while reading the diff, not as a mechanical check:
@@ -1,68 +0,0 @@
-# Examples Python Policy
-
-Prevent new model-specific Python examples without forcing cleanup of the
-existing backlog. Run this check on every PR in quick and full modes.
-
-## Scope
-
-Set the merge base as in the main workflow, then list Python paths introduced
-by the PR, including additions, copies, and renames:
-
-```bash
-BASE="$(git merge-base HEAD origin/main)"
-git diff --name-status --diff-filter=ACR --find-renames --find-copies \
-  "${BASE}"...HEAD -- 'examples/**/*.py'
-```
-
-Inspect destination paths for copies and renames. Existing model-specific files
-that are only modified or deleted are grandfathered and must not be reported.
-
-## Blocking classification
-
-Mark ✗ when a newly introduced Python file does any of the following:
-
- Names or scopes a path to one model, checkpoint, vendor, or model family.
- Hard-codes a model identifier, model-specific prompt, request shape, output
-  conversion, or launch configuration behind an otherwise generic filename.
- Duplicates an existing task/protocol runner to demonstrate one model.
- Adds a per-model helper below a generic task hub such as `text_to_speech/`.
-
-Do not infer safety from a generic filename alone. For example, a file named
-`text_to_audio.py` is still model-specific if it only implements one model's
-contract.
-
-## Acceptable additions
-
-A new Python example may pass when it is a genuinely model-neutral,
-user-facing task or protocol entrypoint. It must accept model selection through
-configuration and keep model-specific behavior out of the script.
-
-Use ⚠ when the file is model-neutral but reusable logic appears misplaced under
-`examples/`. Route code by purpose:
-
-| Code purpose | Preferred location |
-|---|---|
-| Model prompt, defaults, request/output adaptation | `vllm_omni/model_extras/` or production model modules |
-| Runnable model command and validation evidence | Task documentation or `recipes/` |
-| UI or interactive application | `apps/` |
-| Benchmark or evaluator | `benchmarks/` |
-| Regression reproducer | `tests/` |
-| Download, conversion, or setup utility | `tools/` |
-
-If a model cannot use an existing shared runner, report the missing generic
-capability. Do not solve the gap by adding a model-specific Python example.
-
-## Report format
-
-Add one row to the pre-check report:
-
-```text
-Examples policy    ✓ no new Python example paths
-Examples policy    ⚠ generic helper belongs in tools/
-Examples policy    ✗ examples/offline_inference/new_model/end2end.py is model-specific
-Examples policy    ✗ examples/offline_inference/x_to_y_model_name.py is model-specific
-```
-
-For a blocker, name the path, the model-specific behavior, and the appropriate
-shared runner or destination. Do not propose deleting unrelated existing
-examples as part of the PR.
@@ -97,7 +97,6 @@ If docs and live code disagree, verify the code/tests and report the drift.
 | [test-quality-evaluation.md](references/checks/test-quality-evaluation.md) | Tests change, are absent for risky code, or may not exercise production behavior. |
 | [tests-docs-checklist.md](references/checks/tests-docs-checklist.md) | Coverage, CI markers, examples, user docs, or PR evidence need review. |
 | [verification.md](references/checks/verification.md) | Hardware, a server, or a runnable affected path is available for active verification. |
-| [examples-policy.md](../precheck-pr/references/examples-policy.md) | The PR adds, copies, or renames Python under `examples/`; apply the canonical policy shared with `precheck-pr`. |

 ### Delivery and reviewer coordination

@@ -153,13 +152,6 @@ contract they protect or use only the applicable evidence checks.
 Apply every category in [general-checks.md](references/process/general-checks.md) before
 lower-priority comments.

-If the diff census contains an added, copied, or renamed Python path under
-`examples/`, read and apply the canonical
-[examples policy](../precheck-pr/references/examples-policy.md). Treat a new
-model-specific Python example as blocking. Do not flag model-specific example
-debt that the PR only modifies or removes, and do not run the rest of the
-author-oriented `precheck-pr` workflow.
-
 For each changed value or behavior, trace:

 ```text
@@ -11,15 +11,10 @@ build:
    post_checkout:
      - git fetch --unshallow || true
    pre_install:
-      # `pre_install` runs before RTD upgrades pip, and the seeded pip 23.1
-      # rejects the `typing_extensions` wheel torch needs (fixed in pip 23.2).
-      - python -m pip install --upgrade pip
      # Pre-install CPU torch so the subsequent `pip install .[docs]` does not
      # pull ~2 GB of CUDA 13 libs (cuBLAS/cuDNN/NCCL/...) that exceed RTD's
-      # 1 GB memory cap and time out the mkdocs build. Wheel-only: `--index-url`
-      # replaces PyPI, so an sdist fallback cannot reach its build deps. The
-      # quotes keep YAML from reading `:all:` as a mapping key.
-      - "python -m pip install --only-binary=:all: --index-url https://download.pytorch.org/whl/cpu torch"
+      # 1 GB memory cap and time out the mkdocs build.
+      - python -m pip install --index-url https://download.pytorch.org/whl/cpu torch

 mkdocs:
  configuration: mkdocs.yml
@@ -16,7 +16,6 @@ Easy, fast, and cheap omni-modality model serving for everyone
 ---

 *Latest News* 🔥
- [2026/08] [VeRL-Omni](https://github.com/verl-project/verl-omni) `v0.2.0` is released: faster diffusion RL powered by vLLM-Omni (request-level/step-wise batching with FA3), rebuilt Qwen3-Omni multimodal training (DPO & GSPO), plus LTX-2.3, Qwen-Image-Edit support and more. See the [release notes](https://github.com/verl-project/verl-omni/releases/tag/v0.2.0).
 - [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.
@@ -158,6 +158,7 @@ def _build_omni_kwargs(args, quantization=None):
 def _generate_image(omni, args, prompt, seed):
    """Generate a single image and return (PIL.Image, time_seconds, memory_gib)."""
    from vllm_omni.inputs.data import OmniDiffusionSamplingParams
+    from vllm_omni.outputs import OmniRequestOutput
    from vllm_omni.platforms import current_omni_platform

    generator = torch.Generator(device=current_omni_platform.device_type).manual_seed(seed)
@@ -175,7 +176,7 @@ def _generate_image(omni, args, prompt, seed):
    elapsed = time.perf_counter() - start
    peak_mem = torch.accelerator.max_memory_allocated() / (1024**3)

-    req_out = outputs[0] if isinstance(outputs, list) else outputs
+    req_out = OmniRequestOutput.unwrap_result(outputs)
    if not req_out.images:
        raise ValueError("Could not extract image output from result.")
    img = req_out.images[0]
@@ -205,15 +206,15 @@ def _generate_video(omni, args, prompt, seed):
    elapsed = time.perf_counter() - start
    peak_mem = torch.accelerator.max_memory_allocated() / (1024**3)

-    if isinstance(outputs, list) and isinstance(outputs[0], OmniRequestOutput):
-        first = outputs[0]
-    elif isinstance(outputs, OmniRequestOutput):
-        first = outputs
-    else:
-        raise ValueError("Could not extract video frames from output.")
-
-    if hasattr(first, "images"):
-        frames = first.images[0] if first.images else None
+    first = outputs[0]
+    if hasattr(first, "request_output") and isinstance(first.request_output, list):
+        inner = first.request_output[0]
+        if isinstance(inner, OmniRequestOutput) and hasattr(inner, "images"):
+            frames = inner.images[0] if inner.images else None
+        else:
+            frames = inner
+    elif hasattr(first, "images") and first.images:
+        frames = first.images[0]
    else:
        raise ValueError("Could not extract video frames from output.")

@@ -300,7 +300,7 @@ def benchmark(args: argparse.Namespace) -> None:
        output_idx = 0
        for stage_outputs in omni.generate(all_prompts, sampling_params_list, py_generator=True):
            if stage_outputs.final_output_type == "image":
-                request_output = stage_outputs
+                request_output = stage_outputs.request_output
                request_id = getattr(request_output, "request_id", "")

                images = getattr(request_output, "images", [])
@@ -167,11 +167,9 @@ def main() -> None:
    ]
    native_cmd = [
        sys.executable,
-        "examples/offline_inference/text_to_video/text_to_video.py",
+        "examples/offline_inference/text_to_video/text_to_video_lingbot.py",
        "--model",
        args.model,
-        "--model-class-name",
-        "LingBotVideoPipeline",
        "--prompt",
        args.prompt,
        "--output",
@@ -139,34 +139,6 @@ python benchmarks/tts/bench_tts.py \
    --output-dir ./results
 ```

-#### Fixed IndexTTS 2.5 performance sweep
-
-IndexTTS 2.5 is registered in `model_configs.yaml` like the other TTS models.
-Use `--served-model-name` when the server was launched from a local native
-bundle instead of the registry key. The command below reproduces the acceptance
-workload: Seed-TTS Eval EN, `n=500`, concurrency `4/8/16/32`, five warmups, and
-dataset seed 0.
-
-```bash
-python benchmarks/tts/bench_tts.py \
-    --model IndexTeam/IndexTTS-2.5 \
-    --served-model-name /path/to/indextts-2.5 \
-    --task voice_clone --locale en \
-    --dataset-path /path/to/seedtts_testset \
-    --host 127.0.0.1 --port 8092 \
-    --concurrency 4 8 16 32 \
-    --num-prompts 500 --num-warmups 5 \
-    --output-dir ./results/indextts2_5 \
-    -- \
-    --tokenizer /path/to/indextts-2.5/qwen0.6bemo4-merge \
-    --seed 0 --metric-percentiles 50,95,99 --disable-tqdm --save-detailed
-```
-
-The wrapper benchmarks an already-running server. For reproducible quality
-comparisons, add `--request-seed 42`; production-performance runs should omit
-it. The trailing `--seed 0` controls Seed-TTS row selection rather than model
-sampling.
-
 ### 4. Plot a sweep

 ```bash
@@ -84,17 +84,12 @@ def build_bench_args(
    result_filename: str | None,
    extra_cli_args: list[str],
    output_len: int | None = None,
-    served_model_name: str | None = None,
-    num_warmups: int = 2,
-    request_seed: int | None = None,
 ) -> list[str]:
    """Build the ``vllm bench serve --omni`` command for one (task, concurrency) run."""
    dataset_name = _TASK_TO_DATASET[task]
    backend: str = model_cfg["backend"]
    endpoint: str = model_cfg["endpoint"]
-    task_extra_body: dict[str, Any] = dict((model_cfg.get("task_extra_body") or {}).get(task) or {})
-    if request_seed is not None:
-        task_extra_body["seed"] = request_seed
+    task_extra_body: dict[str, Any] = (model_cfg.get("task_extra_body") or {}).get(task) or {}

    # Resolve dataset path
    if dataset_path:
@@ -114,7 +109,7 @@ def build_bench_args(
        "--port",
        str(port),
        "--model",
-        served_model_name or model,
+        model,
        "--backend",
        backend,
        "--endpoint",
@@ -124,7 +119,7 @@ def build_bench_args(
        "--num-prompts",
        str(num_prompts),
        "--num-warmups",
-        str(num_warmups),
+        "2",
        "--percentile-metrics",
        "ttft,e2el,audio_rtf,audio_ttfp,audio_duration,audio_underrun",
    ]
@@ -157,8 +152,7 @@ def build_bench_args(
        if result_filename:
            cmd += ["--result-filename", result_filename]

-    passthrough = extra_cli_args[1:] if extra_cli_args[:1] == ["--"] else extra_cli_args
-    cmd += passthrough
+    cmd += extra_cli_args
    return cmd


@@ -231,12 +225,6 @@ def main() -> None:
    parser.add_argument(
        "--model", required=True, help="HuggingFace model ID (e.g. Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice)"
    )
-    parser.add_argument(
-        "--served-model-name",
-        default=None,
-        help="Model name/path accepted by the running server. Use this when --model selects a registry entry "
-        "but the server was launched from a local bundle path.",
-    )
    parser.add_argument("--task", default="all", help="Task type: voice_clone | default_voice | voice_design | all")
    parser.add_argument("--locale", default="en", choices=["en", "zh"])
    parser.add_argument("--concurrency", type=int, nargs="+", default=[1, 4], metavar="N")
@@ -248,13 +236,6 @@ def main() -> None:
        metavar="N",
        help="Number of prompts per run. If one value, applied to all concurrency levels.",
    )
-    parser.add_argument("--num-warmups", type=int, default=2, help="Warmup requests before each measured run")
-    parser.add_argument(
-        "--request-seed",
-        type=int,
-        default=None,
-        help="Optional model sampling seed added to every request body for reproducible quality runs",
-    )
    parser.add_argument(
        "--dataset-path", default=None, help="Root of seed-tts-eval dataset (required for voice_clone/default_voice)"
    )
@@ -275,9 +256,6 @@ def main() -> None:
    parser.add_argument("extra", nargs=argparse.REMAINDER, help="Extra args passed directly to vllm bench serve")
    args = parser.parse_args()

-    if args.num_warmups < 0:
-        parser.error("--num-warmups cannot be negative")
-
    model_configs = load_model_configs(Path(args.model_configs))
    if args.model not in model_configs:
        known = "\n  ".join(model_configs.keys())
@@ -330,9 +308,6 @@ def main() -> None:
                result_filename=result_filename,
                extra_cli_args=args.extra or [],
                output_len=args.output_len,
-                served_model_name=args.served_model_name,
-                num_warmups=args.num_warmups,
-                request_seed=args.request_seed,
            )
            result = run_one_benchmark(cmd)
            if result is not None:
@@ -7,17 +7,6 @@
 # is tracked here.

 models:
-  IndexTeam/IndexTTS-2.5:
-    supported_tasks: [voice_clone]
-    backend: openai-audio-speech
-    endpoint: /v1/audio/speech
-    trust_remote_code: true
-    task_extra_body:
-      voice_clone:
-        extra_params:
-          lang: en
-          text_normalization: true
-
  # -CustomVoice checkpoints lack speaker_encoder weights, so voice_clone is
  # NOT supported (an attempt raises ValueError from _extract_speaker_embedding
  # at model runtime).  Use -Base for voice_clone.
@@ -1,24 +1,7 @@
-# Target vllm version: v0.27.0 (releases/v0.27.0 branch).
+# Target vllm version: v0.26.0 (releases/v0.26.0 branch).
 # Keep VLLM_BASE_TAG aligned with the target vLLM release on every rebase.
-#
-# Release-boundary form: there is no precompiled-wheel pin here, because the
-# base image already ships exactly the target vLLM. Everything the runtime
-# needs — torch, flashinfer, nvJitLink, numpy — comes from that image as one
-# self-consistent set.
-#
-# This block reappears only when the rebase targets a vLLM commit that is NOT a
-# released tag. During the 0.27 rebase, targeting v0.27.0rc1 on top of the
-# v0.26.0 base image required reinstalling vLLM from wheels.vllm.ai and then
-# hand-repairing the fallout: pinning torch ahead of flashinfer (installing it
-# after bound flashinfer's TVM FFI runtime to the old torch and aborted engine
-# core init), registering libnvJitLink.so.13 with ldconfig, and re-pinning numpy
-# under numba's ceiling. All of it was compensation for a base image and a vLLM
-# that disagreed. At a real release boundary they agree, so none of it is
-# needed. If you pin an untagged commit again, restore the reinstall block —
-# agent/tasks/20_update_wheel_url.sh fails the run rather than letting the
-# Dockerfile silently ship the wrong vLLM.
 ARG VLLM_BASE_IMAGE=vllm/vllm-openai
-ARG VLLM_BASE_TAG=v0.27.0
+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
@@ -1,5 +1,5 @@
 ARG VLLM_ASCEND_IMAGE=quay.io/atlas-ci/vllm-ascend
-ARG VLLM_ASCEND_TAG=v0.27.0
+ARG VLLM_ASCEND_TAG=v0.26.0
 FROM ${VLLM_ASCEND_IMAGE}:${VLLM_ASCEND_TAG}

 # WORKDIR /vllm-workspace/vllm
@@ -1,5 +1,5 @@
 ARG VLLM_ASCEND_IMAGE=quay.io/atlas-ci/vllm-ascend
-ARG VLLM_ASCEND_TAG=v0.27.0-a3
+ARG VLLM_ASCEND_TAG=v0.26.0-a3
 FROM ${VLLM_ASCEND_IMAGE}:${VLLM_ASCEND_TAG}

 # WORKDIR /vllm-workspace/vllm
@@ -1,5 +1,5 @@
 ARG VLLM_ASCEND_IMAGE=quay.io/atlas-ci/vllm-ascend
-ARG VLLM_ASCEND_TAG=v0.27.0
+ARG VLLM_ASCEND_TAG=v0.26.0
 FROM ${VLLM_ASCEND_IMAGE}:${VLLM_ASCEND_TAG}

 ARG APP_DIR=/vllm-workspace/vllm-omni
@@ -1,5 +1,5 @@
 ARG VLLM_ASCEND_IMAGE=quay.io/atlas-ci/vllm-ascend
-ARG VLLM_ASCEND_TAG=v0.27.0-a3
+ARG VLLM_ASCEND_TAG=v0.26.0-a3
 FROM ${VLLM_ASCEND_IMAGE}:${VLLM_ASCEND_TAG}

 ARG APP_DIR=/vllm-workspace/vllm-omni
@@ -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=v0.27.0
+ARG VLLM_VERSION=v0.26.0
 RUN git clone -b ${VLLM_VERSION} https://github.com/vllm-project/vllm
 WORKDIR /workspace/vllm

@@ -108,7 +108,7 @@ 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==3.7.2 && \
+    uv pip install triton-xpu==3.7.1 && \
    uv pip install --no-build-isolation --no-deps .

 CMD ["/bin/bash"]
@@ -144,13 +144,12 @@ WORKDIR /workspace/vllm-omni
 COPY . .

 ENV VLLM_OMNI_TARGET_DEVICE=xpu
-# Keep the upstream oneCCL packages paired with the XPU torch runtime.
 RUN --mount=type=cache,target=/root/.cache/uv \
    uv pip install --no-build-isolation \
    --extra-index-url https://mirrors.ustc.edu.cn/pypi/simple/ \
    ".[dev]" && \
-    uv pip uninstall triton && \
-    uv pip install triton-xpu==3.7.2 --reinstall
+    uv pip uninstall triton oneccl oneccl-devel && \
+    uv pip install triton-xpu==3.7.1 --reinstall

 FROM vllm-omni AS vllm-omni-openai

@@ -40,30 +40,16 @@ nav:
  - Models:
    - models/supported_models.md
  - Features:
-    - Overview: features/README.md
-    - Runtime and Stage Execution:
-      - Execution Modes and Streaming: user_guide/diffusion/execution_modes.md
-      - Sleep Mode: features/sleep_mode.md
-    - Quantization:
-      - 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
-      - AutoRound: user_guide/quantization/autoround.md
-      - msModelSlim: user_guide/quantization/msmodelslim.md
-    - Diffusion Acceleration:
-      - Overview and Compatibility:
-        - Overview: user_guide/diffusion_features.md
-        - Feature Compatibility: user_guide/feature_compatibility.md
-      - CPU Offloading: user_guide/diffusion/cpu_offload.md
+    - Sleep Mode: features/sleep_mode.md
+    - 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
+      - Attention Backends: user_guide/diffusion/attention_backends.md
+      - Frame Interpolation: user_guide/diffusion/frame_interpolation.md
      - Parallelism:
        - Overview: user_guide/diffusion/parallelism/overview.md
        - CFG Parallel: user_guide/diffusion/parallelism/cfg_parallel.md
@@ -72,15 +58,23 @@ nav:
        - 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
-      - Attention Backends: user_guide/diffusion/attention_backends.md
-      - Regional Compilation: user_guide/diffusion/regional_compilation.md
-      - Frame Interpolation: user_guide/diffusion/frame_interpolation.md
-      - Startup and Loading: user_guide/diffusion/startup_and_loading.md
+      - CPU Offloading: user_guide/diffusion/cpu_offload.md
      - LoRA: user_guide/diffusion/lora.md
-    - Experimental:
-      - Session State Manager: features/session_state_manager.md
-  - Integrations:
-    - Custom Diffusion Pipeline: features/custom_pipeline.md
+      - Custom Pipeline: features/custom_pipeline.md
+      - Execution Modes: user_guide/diffusion/execution_modes.md
+    - Quantization:
+      - Quantization: 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
+      - GGUF: user_guide/quantization/gguf.md
+      - AutoRound: user_guide/quantization/autoround.md
+      - msModelSlim: user_guide/quantization/msmodelslim.md
    - ComfyUI: features/comfyui.md
  - Fault Tolerance:
    - Failure Modes: user_guide/fault_injection_reliability_matrix.md
@@ -114,7 +108,6 @@ nav:
          - 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
-      - Quantization: design/feature/quantization.md
      - Diffusion Acceleration:
        - Parallelism:
          - CFG-Parallel: design/feature/cfg_parallel.md
@@ -124,8 +117,9 @@ nav:
          - Sequence Parallel: design/feature/sequence_parallel.md
          - Tensor Parallel: design/feature/tensor_parallel.md
          - VAE Patch Parallelism: design/feature/vae_parallel.md
-        - Attention Optimization:
+        - 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
@@ -163,8 +157,7 @@ nav:
 - CLI Reference: cli
 - Community:
  - Governance: community/governance.md
-  - Contact Us: community/contact_us.md
-  - Community Meetings: community/meetings.md
+  - community/*
  - Slack: https://slack.vllm.ai
  - Blog: https://blog.vllm.ai
  - Forum: https://discuss.vllm.ai
@@ -1,9 +1,5 @@
 # Contact Us

- For bugs and feature requests, search or open a GitHub [issue](https://github.com/vllm-project/vllm-omni/issues).
- For technical and usage questions, join the `sig-omni` channel in [vLLM Slack](https://slack.vllm.ai/) or use the [vLLM Forum](https://discuss.vllm.ai/).
- For coordinating contributions and development, use `sig-omni`, the forum, or join a [community meeting](meetings.md).
-
-## Security Disclosures
-
-Do not share vulnerability details in public GitHub issues, pull requests, discussions, or Slack channels. To arrange a private report, send a direct message to one of the [lead maintainers](governance.md#lead-maintainers) in [vLLM Slack](https://slack.vllm.ai/). If you cannot send a direct message, you may post a detail-free request in `sig-omni` asking a lead maintainer to contact you privately; do not include any vulnerability details in that request.
+- For technical questions and feature requests, please use GitHub [Issues](https://github.com/vllm-project/vllm-omni/issues)
+- For coordinating contributions and development and discussing with other users and developers, please join `sig-omni` channel in our [Slack](https://slack.vllm.ai/) or use the [vLLM Forum](https://discuss.vllm.ai/)
+- For security disclosures, please use GitHub's [Security Advisories](https://github.com/vllm-project/vllm-omni/security/advisories) feature
@@ -52,11 +52,31 @@ Committers have write access and merge rights. They typically have deep expertis
 - [@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

-## Reviewer Routing
+## Diffusion Workload Division

-For path-based reviewer routing, consult [`.github/CODEOWNERS`](https://github.com/vllm-project/vllm-omni/blob/main/.github/CODEOWNERS). Detailed subsystem boundaries and safe-change guidance live in the [module design documents](../design/index.md#module-design-documents).
+This section breaks down diffusion-related responsibilities. If you have PRs touching these areas, please ping the listed owners for review.

-`CODEOWNERS` is operational routing metadata, not a governance or approval policy. Being listed does not grant committer status or merge rights, and it does not mean every listed reviewer must approve a change. The maintainer roster above is authoritative for project roles and merge rights.
+### Runtime-related
+
+- Diffusion models: @RuixiangMa, @wtomin, @xuechendi
+
+### Optimization
+
+- Parallel: @RuixiangMa, @wtomin
+- Attention backend: @ZJY0516
+- Cache: @yuanheng-zhao, @RuixiangMa, @SamitHuang
+- Offload: @yuanheng-zhao
+- Intel GPU support: @xuechendi
+- Quantization: @lishunyang12, @david6666666
+- Kernel fusion / communication-computation: @ZJY0516
+
+### Docs & Test
+
+- Diffusion docs: @RuixiangMa, @wtomin
+
+## Meetings
+
+Committers hold **bi-weekly meetings** to discuss future directions and collaborations of the project.

 ## Committer Nomination Process

@@ -1,8 +0,0 @@
-# Community Meetings
-
-vLLM-Omni hosts developer-facing meetings for Chinese- and English-language audiences. Schedules and access details can change, so use these maintained documents as the source of truth:
-
- [Chinese-language meeting document](https://docs.google.com/document/d/1pdUBiS_7mdOUNDtdwy-9OUf7jsMWN324BIbps5olDME/edit?tab=t.0)
- [English-language meeting document](https://docs.google.com/document/d/19AlD08T8uJ7wqfs8VtYCi0iGMTwlTYVVlAxFZ3S7OIg/edit?tab=t.0)
-
-Each document contains the current schedule, access details, agendas, and past meeting notes. For broader vLLM community events, see the [vLLM events calendar](https://vllm.ai/events/).
@@ -0,0 +1 @@
+# Meetups
@@ -0,0 +1,12 @@
+# Volunteers for Bugfix and CI
+
+We encourage you to check current docs and [issues](https://github.com/vllm-project/vllm-omni/issues) to find possible solutions for your questions. If non of these can solve it, please propose an issue to describe your questions about bug or CI problems for developing.
+
+If you have urgent need for locating and solving bugfix or CI problems, please find community volunteers below.
+
+| Dec 4-Dec 12 | Dec 15-Dec 19 | Dec 22-Dec 26 | Dec 29- Jan 2, 2026| Jan 5-Jan 9 | Jan 12-Jan 16 |
+|----------|----------|----------|----------|----------|----------|
+| <a href="https://github.com/congw729">Conw729</a> | <a href="https://github.com/yinpeiqi">yinpeiqi</a> | <a href="https://github.com/tzhouam">tzhouam</a> | <a href="https://github.com/SamitHuang">SamitHuang</a> | <a href="https://github.com/gcanlin">gcanlin</a> | <a href="https://github.com/natureofnature">natureofnature</a> |
+| <a href="https://github.com/david6666666">david6666666</a> | <a href="https://github.com/R2-Y">R2-Y</a> | <a href="https://github.com/hsliuustc0106">hsliuustc0106</a> | <a href="https://github.com/Gaohan123">Gaohan123</a> | <a href="https://github.com/ZJY0516">ZJY0516</a> | <a href="https://github.com/qibaoyuan">qibaoyuan</a> |
+
+We kindly welcome more contributors to fix bugs and contribute new features!
@@ -3,7 +3,7 @@
 Thank you for your interest in contributing to vLLM-Omni! This document provides guidelines and instructions for contributing.

 !!! note
-    vLLM-Omni hosts developer-facing meetings for Chinese- and English-language audiences. See [Community Meetings](../community/meetings.md) for current schedules, access details, agendas, and past notes.
+    We host weekly developer-facing online meetings to discuss milestones and updates **every Tuesday at 19:30 PDT**. Meeting link as well as the past meeting notes can be found [here](https://tinyurl.com/vllm-omni-meeting).

 ## Getting Started

@@ -122,7 +122,7 @@ Repository skills guide the agent, but they do not replace contributor judgment,
 If you encounter a bug or have a feature request, please search existing issues first to see if it has already been reported. If not, please file a new issue, providing as much relevant information as possible.

 !!! important
-    Do not report suspected security vulnerabilities through a public issue, pull request, discussion, or Slack channel. Follow the [security disclosure instructions](../community/contact_us.md#security-disclosures) to arrange a private report.
+    If you discover a security vulnerability, please report it by creating a GitHub issue with the `security` label.

 ## Pull Requests & Code Reviews

@@ -5,8 +5,6 @@ models. DLO keeps only a small number of DiT blocks on the accelerator and
 streams the remaining blocks from host memory. The distributed backend can
 either shard those host-side weights across an existing parallel group or keep
 the standard loader's rank-local weights and avoid an additional collective.
-The rank-local mmap storage design is tracked in
-[RFC #6195](https://github.com/vllm-project/vllm-omni/issues/6195).

 For user-facing commands, see the
 [distributed layerwise offloading guide](../../user_guide/diffusion/cpu_offload.md)
@@ -18,10 +16,8 @@ DLO is implemented for multi-device diffusion execution. The default
 AllGather path is the primary path for DP and SP deployments. The
 `--dlo-no-use-allgather` path is a rank-local compatibility mode: it is useful
 for standard-loader sharding, workstation bring-up, and systems where an
-additional DLO collective is undesirable. For explicitly mmap-compatible
-models in a TP1, non-HSDP topology, this path shares file-backed DiT checkpoint
-pages across processes and keeps only bounded staging buffers private. TP,
-HSDP, and other models retain standard-loader rank-local host storage.
+additional DLO collective is undesirable, but it does not reduce host weight
+storage across ranks.

 The compatibility matrix below describes the current implementation. The
 unit-level guards are covered, but not every parallelism combination has a
@@ -53,10 +49,6 @@ block's shard is copied to a device buffer and reconstructed with
 `all_gather_into_tensor` on a communication stream while the current block is
 executing.

-Multi-rank AllGather behavior is unchanged. When the effective DLO group size
-is one, the backend now uses rank-local mmap staging; this closes a previous
-loader/backend mismatch that could leave parameters on the `meta` device.
-
 ```text
 Compute:    [Block N]             [Block N+1]          [Block N+2]
 H2D:                      [shard N+1]           [shard N+2]
@@ -76,18 +68,9 @@ requests must take the same execution path at every denoising step.
 ### Rank-local path without DLO AllGather

 With `--dlo-no-use-allgather`, DLO forces its internal offload shard size to
-one and streams complete runtime blocks using H2D copies only. Host storage is
-selected independently from that transfer protocol:
-
- For an explicitly mmap-compatible model with TP1 and no HSDP, each worker
-  retains immutable safetensors views backed by the node's shared OS page
-  cache. Two pinned host buffers, sized to the largest streamed block, stage
-  mmap pages for asynchronous H2D. Checkpoint-to-runtime adapters run while a
-  block is packed, so transformed weights do not become a persistent private
-  model copy.
- With TP, HSDP, online quantization, or a model without mmap support, the
-  regular loader remains responsible for preparing rank-local tensors. This
-  preserves TP/HSDP/quantization loader callbacks.
+one. The regular model loader remains responsible for preparing each rank's
+weights, including TP-local tensors or HSDP-managed parameters. DLO then
+streams those rank-local tensors block by block using H2D copies only.

 This mode means:

@@ -97,26 +80,19 @@ This mode means:
  not shard weights across SP ranks.
 - TP/HSDP/SP collectives, if configured, are not disabled by this flag; only
  DLO's additional weight AllGather is disabled.
- Pure DP deployments with a compatible mmap checkpoint share the streamed
-  DiT's file-backed pages. Other components and fallback paths may still keep
-  rank-private host weights.
+- Pure DP deployments keep a full host-side model copy per rank, subject to
+  shared OS page-cache behavior.
 - The scheduler does not require a synchronized DP request wave for DLO.

-The mmap source remains immutable in both transfer modes. AllGather copies a
-persistent `1 / group_size` shard from it and then closes the mappings;
-rank-local mode keeps the mappings open and copies a complete block through
-bounded staging without adding a collective.
-
-The current implementation packs mmap views into a staging slot synchronously
-on the host; the subsequent pinned H2D transfer overlaps with block compute.
-This keeps host memory bounded but makes CPU packing bandwidth part of the
-rank-local path's performance profile.
+This path is intentionally not implemented by reusing the AllGather mmap
+loader. It relies on the standard loader so model-specific TP/HSDP transforms
+remain intact.

 ## Parallelism compatibility

 | Parallelism | DLO + AllGather | DLO without AllGather |
 |---|---|---|
-| **DP** | Supported primary path. DLO shards host weights across the DP group and can run DP multi-concurrency. | Supported rank-local path. DP replicas remain independent; no DLO weight collective. Compatible TP1 models share file-backed DiT pages across replicas. |
+| **DP** | Supported primary path. DLO shards host weights across the DP group and can run DP multi-concurrency. | Supported rank-local path. DP replicas remain independent; no DLO weight sharding or cross-DP DLO collective. |
 | **SP** | Supported in the implementation. With DP=1, DLO uses the SP group for host-weight sharding; SP still shards sequence/activation work. | SP remains active, but DLO keeps standard-loader rank-local weights and adds no SP weight collective. |
 | **TP > 1** | Unsupported in the DLO mmap path. TP-aware loader callbacks are bypassed, so the backend rejects this configuration when it enters that path. | Standard loading is retained and TP-local tensors can be streamed. This is the intended compatibility path, but it still needs broader model and hardware validation. |
 | **HSDP** | Rejected. HSDP has already sharded parameters, so DLO AllGather would double-shard them. | Accepted by configuration. HSDP owns parameter sharding and its own gathers; DLO only stages rank-local parameters. End-to-end coverage is limited. |
@@ -146,35 +122,11 @@ AllGather DP multi-concurrency requires:
 The no-AllGather path does not impose these DLO-specific synchronized-wave
 requirements.

-The mmap loader is used when the model explicitly declares a compatible
-checkpoint layout. It is independent of AllGather for TP1/non-HSDP execution.
-Models can declare a per-parameter checkpoint-to-runtime adapter; the adapter
-must preserve dtype and element count so it can be applied while packing one
-staging block. MiniMax-H3 uses this mechanism for grouped QKV. Online
-quantization and TP/HSDP no-AllGather execution use the regular loader path.
-
-### Adding mmap support to another model
-
-A new pipeline should opt in only after its ordinary loader contract can be
-represented without materializing a private full-model copy:
-
-1. Populate `weights_sources` with safetensors component sources and prefixes
-   that place every DiT checkpoint key in the pipeline parameter namespace.
-2. If prefixed checkpoint keys already equal runtime parameter names, declare
-   `_supports_mmap_loading = True`. If names differ, provide
-   `_remap_ckpt_key` to map checkpoint names to runtime names (or return
-   `None` for intentionally ignored keys).
-3. If a parameter needs a TP1 layout conversion, attach a callable
-   `mmap_weight_transform` to that parameter during model construction. The
-   transform must preserve dtype and element count because it runs while one
-   bounded block is packed.
-4. Verify that mmap and the regular loader produce the same runtime tensors,
-   that every expected DiT tensor maps successfully, and that TP, HSDP, and
-   online-quantized configurations retain their standard-loader behavior.
-
-Setting `_supports_mmap_loading = True` alone is not sufficient: explicit
-opt-ins fail during initialization if any expected DiT tensor remains on the
-`meta` device after checkpoint mapping.
+The mmap loader is used only by the supported DLO+AllGather path when the
+model has a compatible checkpoint layout. Online quantization is incompatible
+with that sharded mmap path; use `--dlo-no-use-allgather` or disable online
+quantization. Models that do not meet the mmap requirements use the regular
+loader path.

 ## Validation coverage

@@ -184,8 +136,6 @@ Current source-level validation includes:
 - HSDP + DLO without AllGather acceptance at configuration level;
 - TP rejection in the DLO+AllGather mmap path;
 - resident-layer requests requiring no-AllGather;
- rank-local mmap source retention, layout transforms, and bounded host
-  staging;
 - DP request-wave validation for denoising-step compatibility;
 - sharding, double-buffer, AllGather-size, and heterogeneous-block regression
  tests.
@@ -200,8 +150,7 @@ on the target CUDA/NCCL or CANN/HCCL hardware.
  scaling path.
 - Use **SP + DLO AllGather** for long-sequence workloads when DP concurrency is
  not the goal.
- Use **no-AllGather** when replicas must avoid DLO collectives. TP1 models with
-  mmap support can share DiT checkpoint pages; TP/HSDP and unsupported models
-  retain the higher-memory standard-loader path.
+- Use **no-AllGather** to bring up TP or workstation PCIe configurations, with
+  the expectation of higher host-memory use and lower validation confidence.
 - Prefer **HSDP alone** for production HSDP deployments until the combined
  HSDP + DLO no-AllGather path has broader end-to-end coverage.
@@ -11,10 +11,6 @@ in the active navigation.

 ## Feature Design Documents

-For user-facing configuration and current compatibility, see the
-[Features overview](../features/README.md). A design document defines an
-implementation contract; it is not, by itself, a general support claim.
-
 ### Runtime and stage execution

 - [Disaggregated Inference](feature/disaggregated_inference.md)
@@ -34,10 +30,6 @@ implementation contract; it is not, by itself, a general support claim.
 - [Yuanrong Store Connector](feature/omni_connectors/yuanrong_connector.md)
 - [Yuanrong Transfer Engine Connector](feature/omni_connectors/yuanrong_transfer_engine_connector.md)

-### Quantization
-
- [Quantization](feature/quantization.md)
-
 ### Diffusion acceleration

 #### Parallelism
@@ -50,14 +42,14 @@ implementation contract; it is not, by itself, a general support claim.
 - [Tensor Parallel](feature/tensor_parallel.md)
 - [VAE Patch Parallelism](feature/vae_parallel.md)

-#### Attention optimization
-
-The [Diffusion Attention Backends](../user_guide/diffusion/attention_backends.md)
-guide lists all selectable backends and their platform defaults. Skip-Softmax
-is the backend-related optimization with a standalone feature-design contract:
+#### Attention Backends

 - [Skip-Softmax](feature/skip_softmax.md)

+#### Quantization
+
+- [Quantization](feature/quantization.md)
+
 - [Cache-DiT](feature/cache_dit.md)
 - [TeaCache](feature/teacache.md)
 - [Diffusion Continuous Batching](feature/diffusion_continuous_batching.md)
@@ -1,6 +1,6 @@
 # Prometheus Metrics Design

-This document describes how vLLM-Omni exposes Prometheus metrics for multi-stage pipelines, the constraints that shaped the design, and how the pipeline-level metrics coexist with upstream vLLM per-engine metrics.
+This document describes how vLLM-Omni exposes Prometheus metrics for multi-stage pipelines, the constraints that shaped the design, and how the pipeline-level metrics coexist with upstream vLLM per-engine metrics. This iteration covers the text and audio surface; diffusion / image / video families land in a follow-up.

 ## Objectives

@@ -1,67 +0,0 @@
-# Features
-
-Use this section to choose and configure vLLM-Omni runtime and optimization
-features. The navigation follows the active
-[Feature Design](../design/index.md#feature-design-documents) taxonomy whenever
-an implementation contract has a user-facing workflow.
-
-!!! note "User guides, design documents, and recipes"
-
-    User guides explain how to enable a feature and where it is supported.
-    Design documents define internal contracts and do not, by themselves,
-    imply general or production support. For model-specific launch flags and
-    hardware requirements, use the linked examples and validated recipes from
-    [Supported Models](../models/supported_models.md).
-
-## Runtime and Stage Execution
-
-| Goal | User guide | Related design contract |
-| --- | --- | --- |
-| Choose serial, batched, step-wise, or streaming diffusion execution | [Execution Modes and Streaming](../user_guide/diffusion/execution_modes.md) | [Diffusion Continuous Batching](../design/feature/diffusion_continuous_batching.md), [Async Diffusion Output](../design/feature/async_diffusion_output.md) |
-| Reclaim stage memory without restarting the server | [Sleep Mode](sleep_mode.md) | Runtime lifecycle behavior is documented in the user guide |
-
-Some runtime designs are deliberately not promoted as standalone User Guide
-features yet:
-
- [Disaggregated Inference](../design/feature/disaggregated_inference.md) and
-  [OmniConnector implementations](../design/index.md#communication) describe
-  topology and transport contracts. Practical deployment configuration remains
-  under [Pipeline and deploy configurations](../configuration/stage_configs.md).
- [Async Chunk](../design/feature/async_chunk.md) and
-  [Async Omni Output Materialization](../design/feature/omni_async_output_materialization.md)
-  are model- and pipeline-dependent. Use the selected model's deploy
-  configuration and recipe for the supported settings.
- [Automatic Prefix Caching](../design/feature/prefix_caching.md) remains a
-  design-level contract until its user-facing configuration and compatibility
-  surface is consolidated.
-
-## Quantization
-
-Quantization is a cross-model feature rather than a diffusion-only
-optimization. The unified [`quantization_config` guide](../user_guide/quantization/overview.md)
-covers diffusion-only models, multi-stage omni/TTS models, and multi-stage
-diffusion models. Its [design contract](../design/feature/quantization.md)
-defines the shared configuration and backend extension points.
-
-## Diffusion Acceleration
-
-| Goal | User guide | Related design contract |
-| --- | --- | --- |
-| Compare acceleration methods and supported combinations | [Overview](../user_guide/diffusion_features.md), [Feature Compatibility](../user_guide/feature_compatibility.md) | [Diffusion acceleration designs](../design/index.md#diffusion-acceleration) |
-| Move weights between host and device memory | [CPU Offloading](../user_guide/diffusion/cpu_offload.md) | [Distributed Layerwise Offload](../design/feature/distributed_layerwise_offload.md) |
-| Reuse denoising computation | [Cache Acceleration](../user_guide/diffusion/cache_acceleration/cache_dit.md) | [Cache-DiT](../design/feature/cache_dit.md), [TeaCache](../design/feature/teacache.md) |
-| Distribute diffusion work across devices | [Parallelism](../user_guide/diffusion/parallelism/overview.md) | [Parallelism designs](../design/index.md#parallelism) |
-| Select dense, sparse, or quantized attention paths | [Attention Backends](../user_guide/diffusion/attention_backends.md) | [Skip-Softmax](../design/feature/skip_softmax.md) |
-| Compile repeated diffusion regions | [Regional Compilation](../user_guide/diffusion/regional_compilation.md) | User-facing optimization guide |
-| Add generated video frames | [Frame Interpolation](../user_guide/diffusion/frame_interpolation.md) | User-facing extension guide |
-| Reduce diffusion model startup time | [Startup and Loading](../user_guide/diffusion/startup_and_loading.md) | User-facing loading guide |
-| Apply diffusion adapters | [LoRA](../user_guide/diffusion/lora.md) | User-facing extension guide |
-
-The [Pipeline Parallelism guide](../user_guide/diffusion/parallelism/pipeline_parallel.md)
-remains available by direct link, but it is not promoted in the primary User
-Guide navigation until its user-facing support and placement are settled.
-
-## Experimental
-
-[Session State Manager](session_state_manager.md) is opt-in and experimental.
-Its APIs and compatibility may change without notice.
@@ -42,8 +42,7 @@ th {
 | `Wan22VACEPipeline` | Wan2.1-VACE | `Wan-AI/Wan2.1-VACE-1.3B-diffusers`, `Wan-AI/Wan2.1-VACE-14B-diffusers` | ✅︎ | ✅︎ | ✅︎ | ✅︎ | — |
 | `Wan22VACEPipeline` | Wan2.2-VACE | `Pyros13/Wan2.2-VACE-Fun-A14B-Diffusers` | ✅︎ |   |   |   | — |
 | `LTX2Pipeline` | LTX-2 / LTX-2.3 one-stage T2V and I2V | `Lightricks/LTX-2`, `diffusers/LTX-2.3-Diffusers` | ✅︎ | ✅︎ | | | — |
-| `LTX2TwoStagePipeline` | LTX-2 / LTX-2.3 ordinary two-stage T2V and I2V | `Lightricks/LTX-2`, `diffusers/LTX-2.3-Diffusers` + matching Lightricks LoRA and upsampler | ✅︎ | ✅︎ | | | — |
-| `LTX2DistilledPipeline` | LTX-2 / LTX-2.3 full-distilled two-stage T2V and I2V | `rootonchair/LTX-2-19b-distilled`, `diffusers/LTX-2.3-Distilled-Diffusers` + `Lightricks/LTX-2.3` upsampler | ✅︎ | ✅︎ | | | — |
+| `LTX2DistilledPipeline` | LTX-2 distilled two-stage T2V and I2V | `rootonchair/LTX-2-19b-distilled` | ✅︎ | ✅︎ | | | — |
 | `LingBotVideoPipeline` | LingBot-Video dense and MoE T2I, T2V, TI2V | `robbyant/lingbot-video-dense-1.3b`, `robbyant/lingbot-video-moe-30b-a3b` | ✅︎ | | | | — |
 | `MiniMaxH3Pipeline` | MiniMax H3 T2VA, first/last-keyframe FL2VA, and mixed-reference Ref2VA | `MiniMaxAI/MiniMax-H3` | ✅︎ | ✅︎ | | | [Published](https://recipes.vllm.ai/MiniMaxAI/MiniMax-H3) |
 | `DreamZeroPipeline` | DreamZero-DROID | `GEAR-Dreams/DreamZero-DROID` | ✅︎ | ✅︎ | ✅︎ | ✅︎ | — |
@@ -59,7 +58,6 @@ th {
 | `CosyVoice3Model` | CosyVoice3 | `FunAudioLLM/Fun-CosyVoice3-0.5B-2512` | ✅︎ | ✅︎ | | ✅︎ | — |
 | `OmniVoiceModel` | OmniVoice | `k2-fsa/OmniVoice` | ✅︎ | | | | — |
 | `VoxCPM2TalkerForConditionalGeneration` | VoxCPM2 | `openbmb/VoxCPM2` | ✅︎ | | | | — |
-| `DotsTTSForConditionalGeneration` | dots.tts | `rednote-hilab/dots.tts-soar` | ✅︎ | | | | — |
 | `MammothModa2ForConditionalGeneration` | MammothModa2-Preview | `bytedance-research/MammothModa2-Preview` | ✅︎ | ✅︎ | | | — |
 | `MammothModa2ForConditionalGeneration` | MammothModa2-Dev (AR-only image understanding) | `bytedance-research/MammothModa2-Dev` | ✅︎ | | | | — |
 | `Flux2KleinPipeline` | FLUX.2-klein | `black-forest-labs/FLUX.2-klein-4B`, `black-forest-labs/FLUX.2-klein-9B` | ✅︎ | ✅︎ | ✅︎ | ✅︎ | — |
@@ -82,7 +80,6 @@ th {
 | `HiggsAudioV2ForConditionalGeneration` | Higgs-Audio v2 | `bosonai/higgs-audio-v2-generation-3B-base` | ✅︎ | | | | — |
 | `HiggsMultimodalQwen3ForConditionalGeneration` | Higgs-Audio v3 (TTS) | `bosonai/higgs-audio-v3-tts-4b` | ✅︎ | | | | [Published](https://recipes.vllm.ai/bosonai/higgs-audio-v3-tts-4b) |
 | `IndexTTS2TalkerForConditionalGeneration` | IndexTTS-2 | `IndexTeam/IndexTTS-2` | ✅︎ | | | | — |
-| `IndexTTS25TalkerForConditionalGeneration` | IndexTTS-2.5 | native `checkpoints/` bundle | ✅︎ | | | | [Repository](https://github.com/vllm-project/vllm-omni/blob/main/recipes/IndexTeam/IndexTTS-2_5.md) |
 | `NextStep11Pipeline` | NextStep-1.1 | `stepfun-ai/NextStep-1.1` | ✅︎ | ✅︎ | | ✅︎ | — |
 | `MiMoAudioModel` | MiMo-Audio-7B-Instruct | `XiaomiMiMo/MiMo-Audio-7B-Instruct` | ✅︎ | ✅︎ | | | — |
 | `MiMoV2ASRForCausalLM` | MiMo-V2.5-ASR | `XiaomiMiMo/MiMo-V2.5-ASR` | ✅︎ | ✅︎ | | | — |
@@ -10,12 +10,12 @@ A: Refer to [GPU memory calculation and configuration](../configuration/gpu_memo

 > Q: I encounter some bugs or CI problems, which is urgent. How can I solve it?

-A: Search the current [issues](https://github.com/vllm-project/vllm-omni/issues) first. If no existing issue matches a bug or CI failure, open a [bug report](https://github.com/vllm-project/vllm-omni/issues/new?template=400-bug-report.yml) with a minimal reproducer, environment details, and relevant logs. For technical or usage questions, use the `sig-omni` channel in [vLLM Slack](https://slack.vllm.ai/) or the [vLLM Forum](https://discuss.vllm.ai/).
+A: At first, you can check current [issues](https://github.com/vllm-project/vllm-omni/issues) to find possible solutions. If none of these satisfy your demand and it is urgent, please find these [volunteers](https://docs.vllm.ai/projects/vllm-omni/en/latest/community/volunteers/) for help.

 > Q: Does vLLM-Omni support AWQ or any other quantization?

-A: AWQ is available through the inherited vLLM quantization registry, and vLLM-Omni supports several additional methods. Actual support and validation vary by model, component, hardware, and checkpoint. See the [quantization overview](../user_guide/quantization/overview.md) for validated combinations and method-specific guides.
+A: We plan to introduce GGUF FP8 prequantized models and online FP8 quantization in version 0.16.0. Support for other quantization types will follow in future releases. For details, please see our [Q1 quantization roadmap](https://github.com/vllm-project/vllm-omni/issues/1057).

 > Q: Does vLLM-Omni support multimodal streaming input and output?

-A: Yes, for specific models and serving paths. Current capabilities include [streaming video input](../serving/video_stream_api.md), [streaming diffusion output](../user_guide/diffusion/execution_modes.md#streaming-output), and experimental full-duplex real-time audio described in the [project overview](../README.md). Check the model and API documentation because input and output support is model-specific.
+A: Not yet. We already put it on the [Roadmap](https://github.com/vllm-project/vllm-omni/issues/165). Please stay tuned!
@@ -1,23 +1,23 @@
 # Production Metrics

-vLLM-Omni exposes Prometheus metrics via the `/metrics` endpoint on the OpenAI-compatible API server.
+vLLM-Omni exposes Prometheus metrics via the `/metrics` endpoint on the OpenAI-compatible API server. This page covers the text and audio surface; diffusion / image / video metrics are tracked in a follow-up PR.

 ```bash
 vllm-omni serve Qwen/Qwen3-Omni-30B-A3B-Instruct --port 8000 --log-stats
 curl http://localhost:8000/metrics
 ```

-**`--log-stats` is required to populate metric data.** Without the flag, the endpoint still returns `200 OK` but the upstream `vllm:*` wrap is not registered at all, and the 15 `vllm_omni:*` families are registered as placeholders with no data written to them. This keeps the runtime cost essentially zero for deployments that don't need monitoring. With the flag, all ~80 families populate.
+**`--log-stats` is required to populate metric data.** Without the flag, the endpoint still returns `200 OK` but the upstream `vllm:*` wrap is not registered at all, and the 15 `vllm:omni_*` families are registered as placeholders with no data written to them. This keeps the runtime cost essentially zero for deployments that don't need monitoring. With the flag, all ~80 families populate.

 ## Metric Namespaces

 | Prefix | Source | Present when |
 |--------|--------|--------------|
-| `vllm_omni:` | vLLM-Omni orchestrator / audio modality / cross-stage transfer | Pipeline-dependent |
+| `vllm:omni_` | vLLM-Omni orchestrator / audio modality / cross-stage transfer | Pipeline-dependent |
 | `vllm:` | Upstream vLLM engine, wrapped by `OmniPrometheusStatLogger` to expose `{stage, replica}` | Pipeline includes an LLM (AR) stage |
 | `http_` / `process_` | Uvicorn / Python runtime | Always |

-## Pipeline-Level Metrics (`vllm_omni:`)
+## Pipeline-Level Metrics (`vllm:omni_`)

 Defined in `vllm_omni/metrics/prometheus.py`. Track request lifecycle across the full multi-stage pipeline.

@@ -25,66 +25,42 @@ Defined in `vllm_omni/metrics/prometheus.py`. Track request lifecycle across the

 | Metric | Type | Labels | Description |
 |--------|------|--------|-------------|
-| `vllm_omni:num_requests_running` | Gauge | `model_name` | Pipeline-global in-flight requests (dispatched to engine, not yet finalized) |
-| `vllm_omni:num_requests_waiting` | Gauge | `model_name` | Requests waiting in the Orchestrator queue |
-| `vllm_omni:requests_success_total` | Counter | `model_name`, `finished_reason` | Total requests by completion reason. `finished_reason` ∈ {`stop`, `length`, `abort`, ...} mirroring upstream `vllm:request_success_total`; aborts cover client disconnect / cancellation paths in addition to upstream `FinishReason.ABORT` |
+| `vllm:omni_num_requests_running` | Gauge | `model_name` | Pipeline-global in-flight requests (dispatched to engine, not yet finalized) |
+| `vllm:omni_num_requests_waiting` | Gauge | `model_name` | Requests waiting in the Orchestrator queue |
+| `vllm:omni_requests_success_total` | Counter | `model_name`, `finished_reason` | Total requests by completion reason. `finished_reason` ∈ {`stop`, `length`, `abort`, ...} mirroring upstream `vllm:request_success_total`; aborts cover client disconnect / cancellation paths in addition to upstream `FinishReason.ABORT` |

 ### Latency

 | Metric | Type | Labels | Description |
 |--------|------|--------|-------------|
-| `vllm_omni:e2e_request_latency_s` | Histogram | `model_name` | Pipeline-global end-to-end request latency in seconds |
+| `vllm:omni_e2e_request_latency_s` | Histogram | `model_name` | Pipeline-global end-to-end request latency in seconds |

-## Audio Modality Metrics (`vllm_omni:`)
+## Audio Modality Metrics (`vllm:omni_`)

 Emitted at request finalize, except for `audio_ttfp_s` (streaming-hook at the first audio packet) and `audio_underrun_s` / `audio_continuity_ok_total` (streaming finalize, after the chunk stream is exhausted). All carry `{model_name, stage, replica}` plus the listed extra label.

 | Metric | Type | Extra label | Description |
 |--------|------|-------------|-------------|
-| `vllm_omni:audio_ttfp_s` | Histogram | — | Time from request arrival to first audio packet/frame |
-| `vllm_omni:audio_duration_s` | Histogram | — | Audio content duration (`audio_frames / sample_rate`) |
-| `vllm_omni:audio_rtf` | Histogram | — | Real-time factor (`stage_gen_time_s / audio_duration_s`); streaming TTS SLO red line `< 1`; uses `RTF_BUCKETS` |
-| `vllm_omni:audio_frames_total` | Counter | — | Cumulative audio frame count; throughput via `rate()` |
-| `vllm_omni:audio_underrun_s` | Histogram | — | Per-request worst-case player deficit; `> 0` indicates listener heard silent gaps |
-| `vllm_omni:audio_continuity_ok_total` | Counter | `threshold_ms` | Incremented when the request's worst underrun stayed below `threshold_ms` |
-| `vllm_omni:audio_skipped_requests_total` | Counter | `reason` | Silent-loss counter — code2wav rejected malformed codec input and returned `200 OK` with empty audio |
+| `vllm:omni_audio_ttfp_s` | Histogram | — | Time from request arrival to first audio packet/frame |
+| `vllm:omni_audio_duration_s` | Histogram | — | Audio content duration (`audio_frames / sample_rate`) |
+| `vllm:omni_audio_rtf` | Histogram | — | Real-time factor (`stage_gen_time_s / audio_duration_s`); streaming TTS SLO red line `< 1`; uses `RTF_BUCKETS` |
+| `vllm:omni_audio_frames_total` | Counter | — | Cumulative audio frame count; throughput via `rate()` |
+| `vllm:omni_audio_underrun_s` | Histogram | — | Per-request worst-case player deficit; `> 0` indicates listener heard silent gaps |
+| `vllm:omni_audio_continuity_ok_total` | Counter | `threshold_ms` | Incremented when the request's worst underrun stayed below `threshold_ms` |
+| `vllm:omni_audio_skipped_requests_total` | Counter | `reason` | Silent-loss counter — code2wav rejected malformed codec input and returned `200 OK` with empty audio |

 The continuity math comes from `vllm_omni/benchmarks/audio_continuity.py::compute_continuity_stats` so the server-side observation aligns with the bench-side definition.

-## Diffusion Metrics (`vllm_omni:`)
-
-Per-request timing breakdowns for diffusion (image/video) stages. Emitted at request finalize when `stage_metrics.diffusion_metrics` is present. All carry `{model_name, stage, replica}`.
-
-| Metric | Type | Description |
-|--------|------|-------------|
-| `vllm_omni:diffusion_exec_s` | Histogram | DiT forward pass execution time per request in seconds |
-| `vllm_omni:diffusion_exec_per_step_s` | Histogram | DiT forward pass execution time per denoising step in seconds (`exec_s / num_inference_steps`) |
-| `vllm_omni:diffusion_preprocess_s` | Histogram | Diffusion input preprocessing time per request in seconds |
-| `vllm_omni:diffusion_postprocess_s` | Histogram | Diffusion output postprocessing (VAE decode) time per request in seconds |
-
-`diffusion_exec_s` uses `SECONDS_BUCKETS` (50 ms → 300 s); the other three use `SECONDS_FAST_BUCKETS` (1 ms → 60 s). `diffusion_exec_per_step_s` is skipped when `num_inference_steps` is not set.
-
-### Example PromQL
-
-```promql
-# P99 DiT exec time by stage
-histogram_quantile(0.99, rate(vllm_omni:diffusion_exec_s_bucket[5m]))
-
-# Ratio of preprocess to total exec wall time
-rate(vllm_omni:diffusion_preprocess_s_sum[5m])
-  / rate(vllm_omni:diffusion_exec_s_sum[5m])
-```
-
-## Cross-Stage Transfer Metrics (`vllm_omni:`)
+## Cross-Stage Transfer Metrics (`vllm:omni_`)

 Per-physical-transfer histograms tracking the data hop between adjacent stages. Labels `{model_name, from_stage, from_replica, to_stage, to_replica}` let dashboards attribute latency to specific replica edges. `from_replica` / `to_replica` are resolved from the orchestrator's sticky-routing binding (`stage_pool.get_bound_replica_id(request_id)`), so no extra plumbing through `TransferEdgeStats` is needed.

 | Metric | Type | Description |
 |--------|------|-------------|
-| `vllm_omni:transfer_size_bytes` | Histogram | Per-transfer payload size in bytes |
-| `vllm_omni:transfer_tx_s` | Histogram | Sender-side time (serialize + submit to connector) |
-| `vllm_omni:transfer_rx_s` | Histogram | Receiver-side time (recv + deserialize) |
-| `vllm_omni:transfer_in_flight_s` | Histogram | Network in-flight time (TX done → RX recv start) |
+| `vllm:omni_transfer_size_bytes` | Histogram | Per-transfer payload size in bytes |
+| `vllm:omni_transfer_tx_s` | Histogram | Sender-side time (serialize + submit to connector) |
+| `vllm:omni_transfer_rx_s` | Histogram | Receiver-side time (recv + deserialize) |
+| `vllm:omni_transfer_in_flight_s` | Histogram | Network in-flight time (TX done → RX recv start) |

 ## vLLM Engine Metrics (`vllm:`)

@@ -107,9 +83,9 @@ For the full list of upstream metrics, see [the vLLM docs](https://github.com/vl

 | Metric group | Multi-stage LLM (Qwen3-Omni) |
 |---|---|
-| `vllm_omni:` request tracking + latency | With `--log-stats` |
-| `vllm_omni:` audio modality | With `--log-stats`, if pipeline has a talker stage |
-| `vllm_omni:` transfer | With `--log-stats`, if pipeline has ≥ 2 stages |
+| `vllm:omni_` request tracking + latency | With `--log-stats` |
+| `vllm:omni_` audio modality | With `--log-stats`, if pipeline has a talker stage |
+| `vllm:omni_` transfer | With `--log-stats`, if pipeline has ≥ 2 stages |
 | `vllm:` engine metrics (per `(stage, replica)`) | With `--log-stats` |
 | `vllm:` MFU metrics | With `--log-stats --enable-mfu-metrics` |

@@ -118,5 +94,5 @@ For the full list of upstream metrics, see [the vLLM docs](https://github.com/vl
 - All time-bearing metrics use the `_s` suffix (values in seconds). Buckets are `SECONDS_BUCKETS` for e2e / generation-style values and `SECONDS_FAST_BUCKETS` (1 ms → 60 s) for the fine-grained transfer and audio-underrun values.
 - Counters use the `_total` suffix (auto-appended by `prometheus_client`).
 - Sizes use the `_bytes` suffix.
- All omni-specific families are prefixed `vllm_omni:`. The upstream `unregister_vllm_metrics()` function is monkey-patched (see `vllm_omni/patch.py`) to a scoped version that still strips upstream `vllm:*` collectors so multi-engine init within one process does not crash on duplicate registration, but preserves anything prefixed `vllm_omni:`.
- Text and audio first-output use distinct families (`vllm:time_to_first_token_seconds` reused from upstream for text; `vllm_omni:audio_ttfp_s` for audio) rather than a single metric with a `modality` label.
+- All omni-specific families are prefixed `vllm:omni_`. The upstream `unregister_vllm_metrics()` function is monkey-patched (see `vllm_omni/patch.py`) to a scoped version that still strips upstream `vllm:*` collectors so multi-engine init within one process does not crash on duplicate registration, but preserves anything prefixed `vllm:omni_` / `vllm_omni`.
+- Text and audio first-output use distinct families (`vllm:time_to_first_token_seconds` reused from upstream for text; `vllm:omni_audio_ttfp_s` for audio) rather than a single metric with a `modality` label.
@@ -189,10 +189,7 @@ multi-device deployments. In the default mode, each rank stores only
 reconstructed at runtime via **AllGather** on a dedicated communication stream,
 overlapped with computation via a fixed double-buffer scheme. The
 `--dlo-no-use-allgather` mode instead streams standard-loader rank-local
-weights independently and does not shard weights across ranks. For explicitly
-mmap-compatible TP1 models, the same mode can retain node-shared file-backed
-DiT weights and use bounded per-worker host staging instead of a private full
-DiT copy.
+weights independently and does not shard weights across ranks.

 For the implementation design and compatibility matrix, see
 [Distributed Layerwise Offload](../../design/feature/distributed_layerwise_offload.md).
@@ -204,9 +201,8 @@ For the implementation design and compatibility matrix, see
  time, regardless of model size
 - **DP multi-concurrency**: N concurrent requests processed in parallel
  (AllGather only gathers weight shards, which are request-independent)
- **mmap weight storage**: compatible checkpoints use views into the shared OS
-  page cache. AllGather copies persistent rank shards; rank-local mode keeps
-  the views and packs complete blocks through two bounded host staging slots
+- **mmap weight loading**: weights loaded as mmap views pointing to shared OS
+  page cache, eliminating O(dp_size × model_size) RSS during model creation
 - **Platform-agnostic**: works on NVIDIA GPU (CUDA/NCCL) and Ascend NPU
  (CANN/HCCL) via vLLM-Omni's platform abstraction

@@ -228,7 +224,7 @@ vllm serve /path/to/model --omni \
  --enable-distributed-layerwise-offload \
  --data-parallel-size 4

-# Without DLO AllGather (compatible TP1 models use node-shared mmap storage)
+# Without DLO AllGather (each rank streams standard-loader rank-local weights)
 vllm serve /path/to/model --omni \
  --enable-distributed-layerwise-offload \
  --data-parallel-size 4 \
@@ -262,26 +258,20 @@ m = Omni(

 ### How model weights are loaded (mmap path)

-When DLO and an mmap-compatible model are active, the offloader:
+When DLO + AllGather is active, the offloader:

 1. **Saves non-persistent buffers** (e.g. RoPE `inv_freq`, timestep `freqs`)
   from the normally-created transformer
 2. **Converts to meta device** via `to_empty(device="meta")`, releasing random
   initialization weights
 3. **Loads checkpoint weights as mmap views** via `safe_open().get_tensor()`,
-   which return views into the OS page cache without eagerly copying the full
-   tensor into anonymous process memory
-4. **Applies declared per-parameter layout adapters while packing**, when the
-   checkpoint and TP1 runtime layouts differ (for example MiniMax-H3 grouped
-   QKV)
+   which return views into the OS page cache (shared across all ranks, 0 RSS)
+4. **Calls `post_load_weights()`** to apply model-specific dtype conversions
+   (e.g. Cosmos3's `time_embedder` → FP32)
 5. **Restores non-persistent buffers** from saved copies

-With AllGather, each rank copies only its persistent shard and then releases
-the source mappings. Without AllGather, mappings remain open and every worker
-owns two pinned staging slots sized to the largest streamed block. The latter
-shares file-backed pages, not process-local Python objects, and does not alter
-request routing or scheduling. Packing into the host slot is synchronous; the
-pinned H2D copy can overlap with computation.
+This approach requires **zero model-specific code changes** — no pipeline or
+transformer modifications are needed.

 ### OffloadPlan (declarative topology metadata)

@@ -323,7 +313,7 @@ request while AllGather synchronizes only weight shards (request-independent).
  `--dlo-no-use-allgather` or disable online quantization.
 - Tensor Parallel > 1 is rejected in the DLO+AllGather mmap path because TP-aware
  weight-loader callbacks are bypassed. The no-AllGather path retains the
-  standard TP loader instead of rank-local mmap and is experimental.
+  standard TP loader and is experimental.
 - HSDP + AllGather is rejected because it would double-shard parameters. HSDP
  with no-AllGather is accepted at configuration level, but has limited
  end-to-end validation.
@@ -383,7 +373,6 @@ reasoner/generator components inside the model forward pass.
 | StableDiffusion3Pipeline | `stabilityai/stable-diffusion-3.5-medium` | `SD3Transformer2DModel` | - | ✓ | - | `"transformer_blocks"` |
 | Wan22I2VPipeline | `Wan-AI/Wan2.2-I2V-A14B-Diffusers` | `WanTransformer3DModel` | ✓ | ✓ | - | `"blocks"` |
 | Wan22Pipeline | `Wan-AI/Wan2.2-T2V-A14B-Diffusers` | `WanTransformer3DModel` | ✓ | ✓ | - | `"blocks"` |
-| MiniMaxH3Pipeline | `MiniMaxAI/MiniMax-H3` (`FL2VA`) | `MiniMaxH3DiTModel` | - | ✓ | ✓ | `"blocks"` |
 | SoulXSingerPipeline / SoulXSingerSVCPipeline | `Soul-AILab/SoulX-Singer` | `DiffLlama` (`cfm_decoder.model.diff_estimator`) | ✓ | ✓ | - | `"layers"` |
 | BagelPipeline | `ByteDance-Seed/BAGEL-7B-MoT` | `Qwen2MoTModel` | - | ✓ | - | `"layers"`, `"customized modules"` |
 | Cosmos3OmniDiffusersPipeline | `nvidia/Cosmos3-Nano`, `nvidia/Cosmos3-Super` | `Cosmos3VFMTransformer`, `Cosmos3LanguageModel` | ✓ | ✓ | ✓ | `"layers"`, `"gen_layers"` |
@@ -393,9 +382,8 @@ reasoner/generator components inside the model forward pass.
 - Layerwise Offloading requires DiT class to define `_layerwise_offload_blocks_attrs` pointing to transformer blocks
 - Distributed Layerwise Offloading reuses the layerwise block topology, but the
  model and parallelism combination still needs validation. AllGather mode
-  additionally requires a compatible checkpoint/mmap layout. For explicitly
-  compatible TP1 models, no-AllGather uses node-shared mmap sources; other
-  layouts retain the standard loader's rank-local tensors. See the
+  additionally requires a compatible checkpoint/mmap layout; no-AllGather mode
+  retains the standard loader's rank-local tensors. See the
  [DLO design](../../design/feature/distributed_layerwise_offload.md) and
  [Cosmos3 DistOffload recipe](https://github.com/vllm-project/vllm-omni/blob/main/recipes/cosmos3/Cosmos3-DistOffload.md)
  for usage examples.
@@ -1,60 +0,0 @@
-# Diffusion Startup and Loading
-
-Large diffusion models can take several minutes to load at startup. vLLM-Omni
-loads safetensors shards in parallel to reduce this initialization time.
-
-Multi-thread weight loading is enabled by default with four threads. No
-configuration is needed for the default behavior.
-
-## Configuration
-
-| Parameter | CLI flag | Default | Description |
-| --- | --- | --- | --- |
-| `enable_multithread_weight_load` | `--disable-multithread-weight-load` | `True` | Pass the flag to disable multi-thread loading |
-| `num_weight_load_threads` | `--num-weight-load-threads` | `4` | Number of parallel weight-loading threads |
-
-!!! tip
-
-    The default balances startup speed and disk I/O contention. Fast NVMe
-    storage may benefit from more threads, while network storage or hard disks
-    may not.
-
-## Online Serving
-
-```bash
-# Default: multi-thread loading with four threads
-vllm serve Qwen/Qwen-Image --omni --port 8091
-
-# Increase the thread count
-vllm serve Wan-AI/Wan2.2-I2V-A14B-Diffusers --omni \
-  --num-weight-load-threads 8
-
-# Disable multi-thread loading
-vllm serve Qwen/Qwen-Image --omni --disable-multithread-weight-load
-```
-
-## Offline Inference
-
-```python
-from vllm_omni import Omni
-
-# Default: multi-thread loading with four threads
-omni = Omni(model="Qwen/Qwen-Image")
-
-# Increase the thread count
-omni = Omni(
-    model="Wan-AI/Wan2.2-I2V-A14B-Diffusers",
-    num_weight_load_threads=8,
-)
-```
-
-## Reference Benchmarks
-
-The following measurements were collected on NVIDIA H800 hardware. Treat them
-as reference results rather than a guarantee for other storage or hardware
-configurations.
-
-| Model | Sequential loading | Multi-thread loading | Speedup |
-| --- | ---: | ---: | ---: |
-| **Qwen/Qwen-Image** (53.7 GiB) | 168 s | 27 s | **6.2x** |
-| **Wan-AI/Wan2.2-I2V-A14B-Diffusers** (64.5 GiB) | 283 s | 56 s | **5.1x** |
@@ -49,7 +49,7 @@ Parallelism methods distribute computation across GPUs without quality loss (mat

 | Method | Description | Best For |
 |--------|-------------|----------|
-| **[Multi-Thread Weight Loading](diffusion/startup_and_loading.md)** | Loads safetensors shards in parallel using a thread pool | All diffusion models; reduces startup from minutes to seconds |
+| **[Multi-Thread Weight Loading](#multi-thread-weight-loading)** | Loads safetensors shards in parallel using a thread pool | All diffusion models; reduces startup from minutes to seconds |

 **Note:** Some acceleration methods can be combined together for optimized performance. See [Feature Compatibility Table](#feature-compatibility) and [Feature Compatibility Tutorial](feature_compatibility.md) for detailed configuration examples.

@@ -100,9 +100,7 @@ The following tables show which models support each feature:

 - **🔀SP (Ulysses & Ring)**: Includes both Ulysses-SP and Ring-Attention methods
 - ✅ = Fully supported
- ✅* = Supported with the constraint listed below the table
 - ❌ = Not supported
- ❓ = Not verified; not recommended

 > Notes:

@@ -121,7 +119,7 @@ The following tables show which models support each feature:
 | **FLUX.2-dev**           |     ✅     |     ✅      |           ✅           |       ✅        |         ✅         |          ❌          |   ✅    |             ✅             |          ✅           |       ❌        |        ❌         |
 | **GLM-Image**            |     ❌     |     ❌      |           ❌           |       ✅        |         ✅         |          ❌          |   ✅    |             ❌             |          ❌           |       ❌        |        ❌         |
 | **Hidream-I1-Full**        |     ❌     |     ❌      |           ❌           |       ❌        |         ✅         |          ❌          |   ❌    |             ❌             |          ❌           |       ❌        |        ❌         |
-| **HunyuanImage3**        |     ❌     |     ✅      |           ❌           |       ❌        |         ✅         |          ❌          |   ❌    |             ❌             |          ❌           |       ✅        |        ✅*        |
+| **HunyuanImage3**        |     ❌     |     ✅      |           ❌           |       ❌        |         ✅         |          ❌          |   ❌    |             ❌             |          ❌           |       ✅        |        ❌         |
 | **Krea 2**               |     ❌     |     ✅      |           ❌           |       ❌        |         ❌         |          ❌          |   ✅    |             ✅             |      ✅ (decode)      |       ❌        |        ❌         |
 | **LongCat-Image**        |     ✅     |     ✅      |           ✅           |       ✅        |         ✅         |          ❌          |   ❌    |             ✅             |          ❌           |       ❌        |        ❌         |
 | **LongCat-Image-Edit**   |     ✅     |     ✅      |           ✅           |       ✅        |         ✅         |          ❌          |   ❌    |             ✅             |          ❌           |       ❌        |        ❌         |
@@ -147,7 +145,6 @@ The following tables show which models support each feature:
 > 2. `Tongyi-MAI/Z-Image-Turbo` and `SII-GAIR/daVinci-MagiHuman-Base-1080p` are distilled models with minimal NFEs; CFG-Parallel is not necessary.
 > 3. Cosmos3 T2I uses `Cosmos3OmniDiffusersPipeline` with `modalities=["image"]`. Model-level CPU offload swaps the nested UND reasoner and GEN generator pathways; layerwise offload remains available for blockwise GEN/UND offload.
 > 4. Krea 2 currently supports single-GPU inference plus LoRA, Cache-DiT, HSDP, CPU/layerwise offload, and VAE-patch-parallel (decode). TP/SP/CFG-Parallel are not yet wired. The few-step distilled (Turbo) checkpoint uses `is_distilled=true` (fixed timestep shift `mu=1.15`); generate at 2048x2048 by default with `num_inference_steps≈8` and `guidance_scale=0`. The Raw checkpoint uses 1024x1024, `num_inference_steps=28`, and `guidance_scale=4.5`.
-> 5. HunyuanImage3 supports step execution. Multi-request step execution requires `TORCH_SDPA`; see [Diffusion Execution Modes](diffusion/execution_modes.md#step-execution).

 ### VideoGen

@@ -158,16 +155,13 @@ The following tables show which models support each feature:
 | **Wan2.1-VACE**              |     ❌     |     ✅      |           ✅           |       ✅        |         ✅         |         ❌         |   ✅    |             ✅             |      ✅ (decode)      |       ❌        |        ❌         |
 | **LTX-2**                    |     ❌     |     ✅      |           ✅           |       ✅        |         ✅         |         ❌         |   ✅    |             ✅             |      ✅ (decode)      |       ❌        |        ❌         |
 | **LTX-2.3**                  |     ❌     |     ✅      |           ✅           |       ✅        |         ✅         |         ❌         |   ✅    |             ✅             |      ✅ (decode)      |       ❌        |        ❌         |
-| **Helios**                   |     ❌     |     ✅      |           ✅           |       ✅        |         ✅         |         ❌         |   ✅    |             ✅             |          ❌           |       ❌        |        ✅*        |
+| **Helios**                   |     ❌     |     ✅      |           ✅           |       ✅        |         ✅         |         ❌         |   ✅    |             ✅             |          ❌           |       ❌        |        ❌         |
 | **HunyuanVideo-1.5 T2V I2V** |     ❌     |     ✅      |           ✅           |       ✅        |         ✅         |         ❌         |   ✅    |             ✅             |  ✅ (encode/decode)   |       ✅        |        ❌         |
 | **DreamID-Omni**             |     ❌     |     ❌      |           ❌           |       ✅        |         ❌         |         ❌         |   ✅    |             ✅             |          ❌           |       ❌        |        ❌         |
 | **Cosmos3**                  |     ❌     |     ✅      |           ✅           |       ✅        |         ✅         |         ❌         |   ✅    |             ✅             |  ✅ (encode/decode)   |       ✅        |        ❌         |
 | **LongCat-Video-Avatar-1.5** |     ❌     |     ❌      |           ❌           |       ❌        |         ❌         |         ❌         |   ❌    |             ❌             |          ❌           |       ❌        |        ❌         |
 | **MiniMax-H3**               | ✅ (FL2VA) |     ✅      |           ✅           |       ❌        |       ✅ (DiT/TE)  |         ❌         |   ✅    |             ✅             |       ✅ (tile)       |      ✅ (DiT)      |        ❌         |

-> **Step execution note:** Helios supports single-request step execution only;
-> use `max_num_seqs=1`.
-
 **Frame Interpolation Support**

 - **Supported**: Wan2.2 text-to-video, image-to-video, and TI2V pipelines
@@ -196,7 +190,7 @@ The following tables show which models support each feature:
 | **🔀Ring-Attn** | ✅ | ✅ | ✅ | | | | | | | | | | | |
 | **🔀CFG-Parallel** | ✅ | ✅ | ✅ | ✅ | | | | | | | | | | |
 | **🔀Tensor Parallel** | ✅ | ✅ | ✅ | ✅ | ✅ | | | | | | | | | |
-| **🔀HSDP** | ❓ | ❓ | ✅ | ❓ | ✅ | ❌ | | | | | | | | |
+| **🔀HSDP** | ❓ | ❓ | ❓ | ❓ | ❓ | ❌ | | | | | | | | |
 | **🔀Expert Parallel** | ❓ | ❓ | ❓ | ❓ | ❓ | ❓ | ❓ | | | | | | | |
 | **💾CPU Offloading (Layerwise)** | ✅ | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | | | | | | |
 | **💾CPU Offloading (Module-wise)** | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❓ | ❓ | ❌ | | | | | |
@@ -207,43 +201,97 @@ The following tables show which models support each feature:

 !!! info

-    1. HSDP can be combined with Ulysses-SP or CFG-Parallel. Tensor Parallel
-       and HSDP are not compatible; other HSDP combinations in the table
-       remain unverified.
+    1. Tensor Parallel and HSDP are not compatible.
    2. TeaCache and Cache-DiT are not compatible.
    3. CPU Offloading (Layerwise) and CPU Offloading (Module-wise) are not compatible.
-    4. The CPU Offloading (Layerwise) row describes local layerwise offload.
-       Multi-device Distributed Layerwise Offload has a separate topology and
-       compatibility matrix in the [CPU Offloading guide](diffusion/cpu_offload.md#distributed-layerwise-offloading).
-    5. The compatibility matrix uses FP8 as the representative quantization method.
+    4. CPU Offloading (Layerwise) supports single-card for now.
+    5. Using FP8-Quant as an example of qunatization methods.
    6. Step Execution is not compatible with any diffusion cache backend. LoRA is supported, but each scheduled batch must use a single adapter (requests with different `lora_request` or `lora_scale` are kept in separate batches).


 ## Multi-Thread Weight Loading

-The loading guide now lives at [Diffusion Startup and
-Loading](diffusion/startup_and_loading.md). This heading remains so existing
-links to this section continue to work.
+Large diffusion models can take several minutes to load weights at startup (e.g., ~3 min for Qwen-Image, ~5 min for Wan2.2 I2V 14B). Multi-thread weight loading speeds up this process by loading safetensors shards in parallel using a thread pool instead of sequentially.
+
+This optimization is **enabled by default** with 4 threads. No configuration is needed for the default behavior.
+
+### Configuration
+
+| Parameter | CLI Flag | Default | Description |
+|-----------|----------|---------|-------------|
+| `enable_multithread_weight_load` | `--disable-multithread-weight-load` | `True` (enabled) | Pass the flag to disable multi-thread loading |
+| `num_weight_load_threads` | `--num-weight-load-threads` | `4` | Number of threads for parallel weight loading |
+
+!!! tip
+    The default of 4 threads balances speed and disk I/O contention. On fast NVMe storage you may benefit from more threads (e.g., 8). On HDD or network storage, the default of 4 avoids saturating I/O bandwidth.
+
+### Online Serving
+
+```bash
+# Default (multi-thread enabled, 4 threads)
+vllm serve Qwen/Qwen-Image --omni --port 8091
+
+# Custom thread count
+vllm serve Wan-AI/Wan2.2-I2V-A14B-Diffusers --omni --num-weight-load-threads 8
+
+# Disable multi-thread loading
+vllm serve Qwen/Qwen-Image --omni --disable-multithread-weight-load
+```
+
+### Offline Inference
+
+```python
+from vllm_omni import Omni
+
+# Default (multi-thread enabled, 4 threads)
+omni = Omni(model="Qwen/Qwen-Image")
+
+# Custom thread count
+omni = Omni(
+    model="Wan-AI/Wan2.2-I2V-A14B-Diffusers",
+    num_weight_load_threads=8,
+)
+```
+
+### Benchmarks
+
+Measured on NVIDIA H800:
+
+| Model | Before | After | Speedup |
+|-------|--------|-------|---------|
+| **Qwen/Qwen-Image** (53.7 GiB) | 168s | 27s | **6.2x** |
+| **Wan-AI/Wan2.2-I2V-A14B-Diffusers** (64.5 GiB) | 283s | 56s | **5.1x** |

 ## Learn More

-The Diffusion Acceleration navigation groups the remaining guides as follows:
+**Cache Acceleration:**

-| Area | Guide |
-| --- | --- |
-| Compatibility | [Feature Compatibility](feature_compatibility.md) |
-| CPU offloading | [CPU Offloading](diffusion/cpu_offload.md) |
-| Cache acceleration | [TeaCache](diffusion/cache_acceleration/teacache.md), [Cache-DiT](diffusion/cache_acceleration/cache_dit.md) |
-| Parallelism | [Parallelism Overview](diffusion/parallelism/overview.md) |
-| Attention | [Attention Backends](diffusion/attention_backends.md) |
-| Compilation | [Regional Compilation](diffusion/regional_compilation.md) |
-| Video extension | [Frame Interpolation](diffusion/frame_interpolation.md) |
-| Startup | [Startup and Loading](diffusion/startup_and_loading.md) |
-| Adapters | [LoRA](diffusion/lora.md) |
+- **[TeaCache Configuration Guide](diffusion/cache_acceleration/teacache.md)** - Parameter tuning, performance tips, troubleshooting
+- **[Cache-DiT Advanced Guide](diffusion/cache_acceleration/cache_dit.md)** - DBCache, TaylorSeer, SCM techniques and optimization

-Related cross-model and runtime features are documented separately:
+**Parallelism Methods:**

- [Quantization](quantization/overview.md) covers diffusion-only models,
-  multi-stage omni/TTS models, and multi-stage diffusion models.
- [Execution Modes and Streaming](diffusion/execution_modes.md) covers the
-  diffusion runtime, including batching, step execution, and streaming output.
+- **[Parallelism Overview](diffusion/parallelism/overview.md)** - Tensor Parallelism, Sequence Parallelism, CFG Parallelism, Pipeline Parallelism, HSDP, and Expert Parallelism
+
+**Memory Optimization:**
+
+- **[CPU Offload Guide](diffusion/cpu_offload.md)** - Offload model components to CPU, reduce GPU memory usage
+- **[VAE Parallelism Guide](diffusion/parallelism/vae_parallelism.md)** - Distribute VAE decode work across GPUs for high-resolution images and videos
+- **[Quantization](quantization/overview.md)** - Quantization methods for diffusion, multi-stage omni/TTS, and multi-stage diffusion models
+
+**Extensions:**
+
+- **[LoRA Inference Guide](diffusion/lora.md)** - Low-Rank Adaptation for style customization and fine-tuning
+- **[Frame Interpolation Guide](diffusion/frame_interpolation.md)** - Worker-side post-generation video frame interpolation for smoother motion
+
+**Execution Modes:**
+
+- **[Diffusion Execution Modes](diffusion/execution_modes.md)** - Configure request batching, step execution, continuous batching, and streaming output
+
+**Startup Optimization:**
+
+- **[Multi-Thread Weight Loading](#multi-thread-weight-loading)** - Speed up model startup by loading safetensors shards in parallel
+
+**Advanced Topics:**
+
+- **[Feature Compatibility](feature_compatibility.md)** - How to combine multiple features for maximum performance
@@ -12,7 +12,6 @@ For backend selection and SageAttention usage, see the [Diffusion Attention Back
 | Model | Default Resolution | Default Frames | Default Steps | Guidance | VRAM (BF16) |
 |---|---|---|---|---|---|
 | `Wan-AI/Wan2.2-T2V-A14B-Diffusers` | 720x1280 | 81 | 40 | 4.0 | ~60 GiB |
-| `robbyant/lingbot-video-dense-1.3b` / `robbyant/lingbot-video-moe-30b-a3b` | 192x320 | 9 | 2 | 3.0 | ~68 GiB (MoE smoke) |
 | `Lightricks/LTX-2` | 512x768 | 121 | 40 | video 3.0 / audio 7.0 | Model-dependent |
 | `hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v` | 480x832 | 121 | 50 | 6.0 | 1×A100 80GB |
 | `hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v` | 720x1280 | 121 | 50 | 6.0 | FP8 + VAE tiling required |
@@ -36,25 +35,6 @@ python text_to_video.py \
  --output t2v_out.mp4
 ```

-### LingBot-Video
-
-The shared runner recognizes both official LingBot checkpoint IDs and selects
-the native `LingBotVideoPipeline` automatically:
-
-```bash
-python text_to_video.py \
-  --model robbyant/lingbot-video-dense-1.3b \
-  --prompt "a robotic arm picks up a red block" \
-  --height 192 --width 320 --num-frames 9 --num-inference-steps 2 \
-  --guidance-scale 3.0 --flow-shift 3.0 --fps 24 \
-  --output lingbot_t2v.mp4
-```
-
-Use `robbyant/lingbot-video-moe-30b-a3b` for the MoE checkpoint. For local
-paths without `lingbot` in the directory name, add
-`--model-class-name LingBotVideoPipeline`. Model-only options use
-`--extra-body`, for example `'{"batch_cfg": true, "output_type": "np"}'`.
-
 ### LTX-2

 ```bash
@@ -80,14 +80,12 @@ warmup_quack_fp8([(14040, 2048, 6144), (14040, 2048, 2048)])

 ## Model Type Support

-### Diffusion Models
+### Diffusion Model (Qwen-Image, Wan2.2)

 | Model | HF models | Online | Pre-calibrated | Recommendation | `ignored_layers` | Text-Encoder quantization |
 |-------|-----------|:-------:|:------:|----------------|------------------|------------------|
 | Qwen-Image | `Qwen/Qwen-Image`, `Qwen/Qwen-Image-2512` | Yes | Yes | Skip sensitive image-stream MLPs when quality regresses | `img_mlp` | |
 | Wan2.2 | Wan2.2 diffusion pipelines | Not validated | Not validated | Validate against BF16 before documenting as supported | TBD | |
-| LTX-2 | `Lightricks/LTX-2`, `rootonchair/LTX-2-19b-distilled` | Yes | Not validated | Transformer only; use dynamic phase LoRA for ordinary two-stage | None | |
-| LTX-2.3 | `diffusers/LTX-2.3-Diffusers`, `diffusers/LTX-2.3-Distilled-Diffusers` | Yes | Not validated | Transformer only; use dynamic phase LoRA for ordinary two-stage | None | |
 | Z-Image | `Tongyi-MAI/Z-Image-Turbo` | Yes | Yes | All layers | None | ✅︎ |
 | FLUX.1 | `black-forest-labs/FLUX.1-dev`, `black-forest-labs/FLUX.1-schnell` | Yes | Yes | All layers | None | |
 | FLUX.2-dev | `black-forest-labs/FLUX.2-dev` | Yes | Not validated | All layers | None | ✅︎ |
@@ -12,20 +12,20 @@ For the internal architecture and backend extension points, see the

 | Mode | Guide | Description | Methods |
 |------|-------|-------------|---------|
-| Online quantization | [Online Quantization](online.md) | vLLM-Omni computes quantized weights and scales while loading the model. | FP8 W8A8, Int8 W8A8, BitsAndBytes W4, MXFP8 W8A8, MXFP4 W4A4 |
+| Online quantization | [Online Quantization](online.md) | vLLM-Omni computes quantized weights and scales while loading the model. | FP8 W8A8, Int8 W8A8, MXFP8 W8A8, MXFP4 W4A4 |
 | Runtime attention quantization | [Quantized KV Cache](quantized_kvcache.md) | vLLM-Omni dynamically quantizes eligible diffusion Flash Attention tensors during inference. | FP8 FA |
-| Pre-quantized checkpoints | Method-specific guides | The checkpoint or an offline quantizer provides quantized weights and scales before serving. | ModelOpt, AutoRound, msModelSlim, serialized Int8, offline MXFP8, offline MXFP4 DualScale |
+| Pre-quantized checkpoints | Method-specific guides | The checkpoint or an offline quantizer provides quantized weights and scales before serving. | ModelOpt, GGUF, AutoRound, msModelSlim, serialized Int8, offline MXFP8, offline MXFP4 DualScale |

 ## Hardware Support

-| Device | FP8 W8A8 | Int8 W8A8 | BitsAndBytes W4 | ModelOpt | MXFP8 W8A8 | MXFP4 W4A4 | AutoRound | msModelSlim |
-|--------|----------|-----------|------------------|----------|------------|------------|-----------|-------------|
-| NVIDIA Blackwell GPU (SM 100+) | ✅ | ✅ | ✅ | ✅ | ⭕ | ⭕ | ✅ | ❌ |
-| NVIDIA Ada/Hopper GPU (SM 89+) | ✅ | ✅ | ✅ | ✅ | ⭕ | ⭕ | ✅ | ❌ |
-| NVIDIA Ampere GPU (SM 80+) | ✅ | ✅ | ✅ | ⭕ | ⭕ | ⭕ | ✅ | ❌ |
-| AMD ROCm | ⭕ | ⭕ | ❌ | ⭕ | ⭕ | ✅ | ⭕ | ❌ |
-| Intel XPU | ⭕ | ⭕ | ❌ | ⭕ | ⭕ | ⭕ | ✅ | ❌ |
-| Ascend NPU | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ✅ |
+| Device | FP8 W8A8 | Int8 W8A8 | ModelOpt | MXFP8 W8A8 | MXFP4 W4A4 | GGUF | AutoRound | msModelSlim |
+|--------|----------|-----------|----------|------------|------------|------|-----------|-------------|
+| NVIDIA Blackwell GPU (SM 100+) | ✅ | ✅ | ✅ | ⭕ | ⭕ | ✅ | ✅ | ❌ |
+| NVIDIA Ada/Hopper GPU (SM 89+) | ✅ | ✅ | ✅ | ⭕ | ⭕ | ✅ | ✅ | ❌ |
+| NVIDIA Ampere GPU (SM 80+) | ✅ | ✅ | ⭕ | ⭕ | ⭕ | ✅ | ✅ | ❌ |
+| AMD ROCm | ⭕ | ⭕ | ⭕ | ⭕ | ✅ | ⭕ | ⭕ | ❌ |
+| Intel XPU | ⭕ | ⭕ | ⭕ | ⭕ | ⭕ | ⭕ | ✅ | ❌ |
+| Ascend NPU | ❌ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ |

 Legend: `✅` supported, `❌` unsupported, `⭕` not verified in this
 guide. FP8 on Ampere may use a weight-only path where available.
@@ -43,10 +43,10 @@ otherwise.
 |--------|-------|------|----------------|--------|
 | FP8 W8A8 | [FP8](fp8.md) | Online W8A8 or checkpoint FP8 | Qwen-Image; Wan2.2 is not validated | Validated for Qwen-Image family and other DiT models |
 | Int8 W8A8 | [Int8](int8.md) | Online or serialized W8A8 | Qwen-Image; Wan2.2 is not validated | Validated for Qwen-Image and Z-Image |
-| BitsAndBytes W4 | [BitsAndBytes](bitsandbytes.md) | Online W4 weight-only | Z-Image-Turbo; Qwen-Image/Wan2.2 not validated | Validated for Z-Image-Turbo |
 | ModelOpt | [ModelOpt](modelopt.md) | Pre-quantized FP8 checkpoints | Qwen-Image, Z-Image, FLUX.2, HunyuanImage-3.0 | Validated for ModelOpt FP8 diffusion checkpoints |
 | MXFP8 W8A8 | [MXFP8](mxfp8.md) | Online W8A8 or offline pre-quantized | Wan2.2-T2V-A14B, I2V-A14B, TI2V-5B | Ascend NPU only; validated for Wan2.2 |
 | MXFP4 W4A4 | [MXFP4](mxfp4.md) | `mxfp4`: online single-scale only; `mxfp4_dualscale`: online or offline dual-scale (offline recommended) | Wan2.2-T2V-A14B, I2V-A14B | Ascend NPU only; validated for Wan2.2 A14B cascade models; TI2V-5B not supported; offline `mxfp4_dualscale` uses calibrated `mul_scale` for best accuracy |
+| GGUF | [GGUF](gguf.md) | Pre-quantized transformer weights | Qwen-Image | Validated where a model-specific GGUF adapter exists |
 | AutoRound | [AutoRound](autoround.md) | Pre-quantized W4A16 checkpoints | FLUX.1-dev; Qwen-Image/Wan2.2 not validated | Checkpoint-driven |
 | msModelSlim | [msModelSlim](msmodelslim.md) | Pre-quantized Ascend checkpoints | Wan2.2 recipe; HunyuanImage-3.0 inference target | Ascend/NPU path |

@@ -61,9 +61,9 @@ in BF16 unless the model guide explicitly adds support.
 |--------|-------|-------|----------------|--------|
 | ModelOpt | [ModelOpt](modelopt.md) | Thinker or language-model checkpoint config | Qwen3-Omni thinker | ModelOpt checkpoint path |
 | Int8 | [Int8](int8.md) | Not currently validated for omni/TTS stages | Qwen3-Omni, Qwen3-TTS | Not validated |
-| BitsAndBytes | [BitsAndBytes](bitsandbytes.md) | Not currently validated for omni/TTS stages | Qwen3-Omni, Qwen3-TTS | Not validated |
 | MXFP8 | [MXFP8](mxfp8.md) | Not currently validated for omni/TTS stages | Qwen3-Omni, Qwen3-TTS | Not validated |
 | MXFP4 | [MXFP4](mxfp4.md) | Not currently validated for omni/TTS stages | Qwen3-Omni, Qwen3-TTS | Not validated |
+| GGUF | [GGUF](gguf.md) | Not currently validated for omni/TTS stages | Qwen3-Omni, Qwen3-TTS | Not validated |
 | AutoRound | [AutoRound](autoround.md) | Thinker or language-model checkpoint config | Qwen2.5-Omni, Qwen3-Omni | Supported through AutoRound checkpoints |
 | msModelSlim | [msModelSlim](msmodelslim.md) | Not currently validated for omni/TTS stages | Qwen3-Omni, Qwen3-TTS | Not validated |

@@ -76,10 +76,10 @@ attached to the intended stage rather than applied globally.
 |--------|-------|-------|----------------|--------|
 | FP8 | [FP8](fp8.md) | Stage-specific DiT or transformer module | BAGEL, GLM-Image | Requires model-specific validation |
 | Int8 | [Int8](int8.md) | Stage-specific DiT or transformer module | BAGEL, GLM-Image | Requires model-specific validation |
-| BitsAndBytes | [BitsAndBytes](bitsandbytes.md) | Stage-specific transformer or DiT module | BAGEL, GLM-Image | Not validated |
 | ModelOpt | [ModelOpt](modelopt.md) | Checkpoint-defined diffusion stage | BAGEL, GLM-Image | Requires model-specific validation |
 | MXFP8 | [MXFP8](mxfp8.md) | Stage-specific DiT or transformer module | BAGEL, GLM-Image | Not validated |
 | MXFP4 | [MXFP4](mxfp4.md) | Stage-specific DiT or transformer module | BAGEL, GLM-Image | Not validated |
+| GGUF | [GGUF](gguf.md) | Stage-specific transformer weights | BAGEL, GLM-Image | No validated adapter listed |
 | AutoRound | [AutoRound](autoround.md) | Checkpoint-defined stage | BAGEL, GLM-Image | No validated checkpoint listed |
 | msModelSlim | [msModelSlim](msmodelslim.md) | Ascend-generated stage weights | GLM-Image | Requires model-specific adaptation |

@@ -106,7 +106,7 @@ config = build_quant_config({

 | Component | Default quantized? | Notes |
 |-----------|--------------------|-------|
-| Diffusion transformer | Yes | Primary target for FP8, Int8, BitsAndBytes, ModelOpt, MXFP8, MXFP4, AutoRound, and msModelSlim |
+| Diffusion transformer | Yes | Primary target for FP8, Int8, ModelOpt, MXFP8, MXFP4, GGUF, AutoRound, and msModelSlim |
 | Text encoder | No | Keep BF16 unless a method-specific guide documents support |
 | VAE | No | Keep BF16; storage-only paths are method-specific |
 | Scheduler/tokenizer | No | Loaded from the base model repository |
@@ -139,8 +139,8 @@ from vllm_omni.quantization import build_quant_config

 build_quant_config("fp8")
 build_quant_config({"method": "fp8", "activation_scheme": "static"})
-build_quant_config("bitsandbytes")
 build_quant_config("auto-round", bits=4, group_size=128)
+build_quant_config({"method": "gguf", "gguf_model": "/path/to/model.gguf"})
 build_quant_config({"transformer": {"method": "fp8"}, "vae": None})
 build_quant_config(None)
 ```
@@ -131,7 +131,7 @@ def build_sampling_params() -> list[SamplingParams]:


 def save_stage_output(stage_outputs, output_dir: str) -> None:
-    output = stage_outputs
+    output = stage_outputs.request_output
    request_id = output.request_id
    if stage_outputs.final_output_type == "text":
        text_output = output.outputs[0].text
@@ -59,7 +59,7 @@ def main(args):

    omni = Omni(
        model=args.model_name,
-        deploy_config=args.deploy_config,
+        stage_configs_path=args.stage_configs_path,
        log_stats=args.log_stats,
        stage_init_timeout=args.stage_init_timeout,
        batch_timeout=args.batch_timeout,
@@ -104,7 +104,7 @@ def main(args):
    os.makedirs(output_dir, exist_ok=True)

    for stage_outputs in omni_outputs:
-        output = stage_outputs
+        output = stage_outputs.request_output
        if stage_outputs.final_output_type == "text":
            request_id = output.request_id
            text_output = output.outputs[0].text
@@ -167,10 +167,10 @@ def parse_args():
        help="Sampling rate for audio loading (default: 16000).",
    )
    parser.add_argument(
-        "--deploy-config",
+        "--stage-configs-path",
        type=str,
        default=None,
-        help="Path to a deploy config YAML file.",
+        help="Path to stage configs YAML file.",
    )
    parser.add_argument(
        "--log-stats",
@@ -228,7 +228,7 @@ async def main():
    if not outputs:
        raise ValueError("No output produced from omni.generate()")

-    first_out = outputs[0]
+    first_out = outputs[0].request_output
    req_out: OmniRequestOutput = first_out

    # Verify trajectory data (from custom pipeline)
@@ -741,7 +741,7 @@ def make_mmu_prompt(

 def iter_mm_outputs(outputs: list[Any]):
    for omni_out in outputs:
-        req_out = omni_out
+        req_out = getattr(omni_out, "request_output", None)
        req_list = req_out if isinstance(req_out, list) else [req_out]
        for item in req_list:
            if item is None:
@@ -876,7 +876,7 @@ def parse_args(repo_root: Path) -> argparse.Namespace:
    parser.add_argument("--task", type=str, required=True, choices=TASK_CHOICES)
    parser.add_argument("--model", type=str, required=True, help="HF repo id or local model directory.")
    parser.add_argument(
-        "--deploy-config",
+        "--stage-config-path",
        type=str,
        default=str(repo_root / "vllm_omni/deploy/dynin_omni.yaml"),
        help="Path to deploy config yaml.",
@@ -1394,8 +1394,8 @@ def main() -> None:

    from vllm_omni.entrypoints.omni import Omni

-    deploy_config_path = str(Path(args.deploy_config).expanduser())
-    omni = Omni(model=model_source, deploy_config=deploy_config_path, dtype=args.dtype)
+    stage_config_path = str(Path(args.stage_config_path).expanduser())
+    omni = Omni(model=model_source, stage_configs_path=stage_config_path, dtype=args.dtype)
    sampling_params_list = [
        SamplingParams(max_tokens=int(args.max_tokens_per_stage), temperature=0.0, top_p=1.0, detokenize=False)
        for _ in range(omni.num_stages)
@@ -73,6 +73,7 @@ def parse_args():
    )
    parser.add_argument("--sys-type", type=str, default=None, help="Override system prompt type.")
    parser.add_argument("--deploy-config", type=str, default=None, help="Custom deploy YAML path.")
+    parser.add_argument("--stage-configs-path", type=str, default=None, help="Custom legacy stage config YAML path.")
    parser.add_argument(
        "--stream",
        action="store_true",
@@ -145,7 +146,13 @@ def main():
    else:
        bot_task = args.bot_task

-    deploy_config = args.deploy_config or _MODALITY_DEFAULT_DEPLOY_CONFIG[args.modality]
+    if args.deploy_config is not None and args.stage_configs_path is not None:
+        raise ValueError("--deploy-config and --stage-configs-path are mutually exclusive.")
+
+    deploy_config = args.deploy_config
+    stage_configs_path = args.stage_configs_path
+    if deploy_config is None and stage_configs_path is None:
+        deploy_config = _MODALITY_DEFAULT_DEPLOY_CONFIG[args.modality]

    omni_kwargs = {
        "model": args.model,
@@ -161,7 +168,10 @@ def main():

    if additional_config is not None:
        omni_kwargs["additional_config"] = additional_config
-    omni_kwargs["deploy_config"] = deploy_config
+    if deploy_config is not None:
+        omni_kwargs["deploy_config"] = deploy_config
+    else:
+        omni_kwargs["stage_configs_path"] = stage_configs_path

    omni = Omni(**omni_kwargs)

@@ -262,7 +272,10 @@ def main():
    print(f"  Modality: {args.modality}")
    print(f"  Prompt task: {task}")
    print(f"  Bot task: {bot_task}")
-    print(f"  Deploy config: {deploy_config}")
+    if deploy_config is not None:
+        print(f"  Deploy config: {deploy_config}")
+    else:
+        print(f"  Stage config: {stage_configs_path}")
    print(f"  Num stages: {omni.num_stages}")
    if args.modality in ("text2img", "img2img"):
        print(f"  Inference steps: {args.steps}")
@@ -290,7 +303,7 @@ def main():
    )
    img_idx = 0
    for req_output in omni_outputs:
-        ro = req_output
+        ro = getattr(req_output, "request_output", None)
        stage_id = getattr(req_output, "stage_id", None)

        # AR stage text — each CompletionOutput.text is already a delta when
@@ -621,7 +621,7 @@ def main():
        images = getattr(output, "images", None)
        if images:
            break
-        req_out = output
+        req_out = getattr(output, "request_output", None)
        images = getattr(req_out, "images", None) if req_out is not None else None
        if images:
            break
@@ -63,9 +63,7 @@ from vllm_omni.inputs.data import OmniDiffusionSamplingParams
 from vllm_omni.lora.request import LoRARequest
 from vllm_omni.lora.utils import stable_lora_int_id
 from vllm_omni.model_extras import (
-    build_image_to_video_prompt as build_model_image_to_video_prompt,
-)
-from vllm_omni.model_extras import (
+    build_image_to_video_prompt,
    get_extra_body_params,
    get_model_class_name,
 )
@@ -87,22 +85,6 @@ def parse_json_object(value: str, flag_name: str = "argument") -> dict[str, Any]
 parse_profiler_config = functools.partial(parse_json_object, flag_name="--profiler-config")


-def build_image_to_video_prompt(
-    prompt: str,
-    negative_prompt: str | None,
-    media_inputs: dict[str, Any],
-) -> dict[str, Any]:
-    """Build the canonical request envelope for the shared I2V example."""
-    result: dict[str, Any] = {
-        "prompt": prompt,
-        "modalities": ["video"],
-        "multi_modal_data": media_inputs,
-    }
-    if negative_prompt is not None:
-        result["negative_prompt"] = negative_prompt
-    return result
-
-
 def parse_args() -> argparse.Namespace:
    parser = argparse.ArgumentParser(
        description=(
@@ -584,13 +566,10 @@ def main():
        # Preserve the historical empty-prompt behavior for non-LTX examples.
        negative_prompt = ""
    prompt_dict = build_image_to_video_prompt(
+        model_class_name=model_class_name,
        prompt=args.prompt,
        negative_prompt=negative_prompt,
        media_inputs=media_inputs,
-    )
-    prompt_dict = build_model_image_to_video_prompt(
-        model_class_name=model_class_name,
-        prompt=prompt_dict,
        height=height,
        width=width,
        num_frames=num_frames,
@@ -646,8 +625,8 @@ def main():
        if frames.multimodal_output and "audio" in frames.multimodal_output:
            audio = frames.multimodal_output["audio"]
            audio_sample_rate = frames.multimodal_output.get("audio_sample_rate", audio_sample_rate)
-        if frames.is_pipeline_output and frames is not None:
-            inner_output = frames
+        if frames.is_pipeline_output and frames.request_output is not None:
+            inner_output = frames.request_output
            if isinstance(inner_output, OmniRequestOutput):
                if inner_output.multimodal_output and "audio" in inner_output.multimodal_output:
                    audio = inner_output.multimodal_output["audio"]
@@ -1,117 +0,0 @@
-"""
-Offline inference example: MammothModa2 image summarization (single AR stage).
-
-Example:
-  uv run python examples/offline_inference/mammothmodal2_preview/run_mammothmoda2_image_summarize.py \
-    --model path/to/MammothModa2-Preview \
-    --deploy-config vllm_omni/deploy/mammoth_moda2_ar.yaml \
-    --image /path/to/input.jpg \
-    --question "Please summarize the content of this image."
-"""
-
-from __future__ import annotations
-
-import argparse
-import os
-
-from PIL import Image
-from vllm import SamplingParams
-from vllm.multimodal.image import convert_image_mode
-
-from vllm_omni import Omni
-
-DEFAULT_SYSTEM = "You are a helpful assistant."
-DEFAULT_QUESTION = "Please summarize the content of this image."
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="MammothModa2 image summarization (offline, AR only).")
-    parser.add_argument("--model", type=str, required=True, help="Path to model directory or model id.")
-    parser.add_argument("--deploy-config", type=str, required=True, help="Path to deploy config yaml.")
-    parser.add_argument("--image", type=str, required=True, help="Path to input image.")
-    parser.add_argument("--question", type=str, default=DEFAULT_QUESTION, help="Question/instruction for the model.")
-    parser.add_argument("--system", type=str, default=DEFAULT_SYSTEM, help="System prompt.")
-    parser.add_argument(
-        "--max-tokens",
-        type=int,
-        default=512,
-        help="Max new tokens to generate.",
-    )
-    parser.add_argument("--temperature", type=float, default=0.2)
-    parser.add_argument("--top-p", type=float, default=0.9)
-    parser.add_argument("--seed", type=int, default=42)
-    parser.add_argument("--trust-remote-code", action="store_true")
-    parser.add_argument(
-        "--enable-diffusion-pipeline-profiler",
-        action="store_true",
-        help="Enable diffusion pipeline profiler to display stage durations.",
-    )
-    return parser.parse_args()
-
-
-def build_prompt(system: str, question: str) -> str:
-    return (
-        f"<|im_start|>system\n{system}<|im_end|>\n"
-        "<|im_start|>user\n"
-        "<|vision_start|><|image_pad|><|vision_end|>"
-        f"{question}<|im_end|>\n"
-        "<|im_start|>assistant\n"
-    )
-
-
-def main() -> None:
-    args = parse_args()
-
-    if not os.path.exists(args.image):
-        raise FileNotFoundError(f"Image file not found: {args.image}")
-
-    pil_image = Image.open(args.image)
-    image_data = convert_image_mode(pil_image, "RGB")
-    prompt = build_prompt(args.system, args.question)
-
-    omni = Omni(
-        model=args.model,
-        deploy_config=args.deploy_config,
-        trust_remote_code=args.trust_remote_code,
-        enable_diffusion_pipeline_profiler=args.enable_diffusion_pipeline_profiler,
-    )
-    try:
-        sp = SamplingParams(
-            temperature=float(args.temperature),
-            top_p=float(args.top_p),
-            top_k=-1,
-            max_tokens=int(args.max_tokens),
-            seed=int(args.seed),
-            detokenize=True,
-        )
-        # NOTE: omni.generate() returns a Generator[OmniRequestOutput, None, None].
-        # Consume it inside the try block so the worker isn't closed early.
-        outputs = list(
-            omni.generate(
-                [
-                    {
-                        "prompt": prompt,
-                        "multi_modal_data": {"image": image_data},
-                        "additional_information": {"omni_task": ["chat"]},
-                    }
-                ],
-                [sp],
-            )
-        )
-    finally:
-        omni.close()
-
-    lines: list[str] = []
-    for stage_outputs in outputs:
-        ro = stage_outputs
-        text = ro.outputs[0].text if getattr(ro, "outputs", None) else str(ro)
-        lines.append(f"request_id: {getattr(ro, 'request_id', 'unknown')}\n")
-        lines.append("answer:\n")
-        lines.append(text.strip() + "\n")
-        lines.append("\n")
-
-    print("\n".join(lines))
-
-
-if __name__ == "__main__":
-    main()
@@ -297,7 +297,7 @@ def main(args):
    os.makedirs(output_dir, exist_ok=True)

    for stage_outputs in omni_outputs:
-        output = stage_outputs
+        output = stage_outputs.request_output
        if stage_outputs.final_output_type == "text":
            request_id = output.request_id
            text_output = output.outputs[0].text
@@ -347,7 +347,7 @@ def main(args):
        omni.start_profile(stages=args.profiler_stages)

    for stage_outputs in omni.generate(prompts, sampling_params_list):
-        output = stage_outputs
+        output = stage_outputs.request_output
        if stage_outputs.final_output_type == "text":
            request_id = output.request_id
            text_output = output.outputs[0].text
@@ -452,6 +452,12 @@ def parse_args():
        default=None,
        help="Path to a .txt file with one prompt per line (preferred).",
    )
+    parser.add_argument(
+        "--stage-configs-path",
+        type=str,
+        default=None,
+        help="Path to a stage configs file (deprecated; prefer --deploy-config).",
+    )
    parser.add_argument(
        "--deploy-config",
        type=str,
@@ -11,13 +11,6 @@ Examples:
    python examples/offline_inference/minimax_h3/dlo_dp2_lifecycle.py \
        --model /path/to/MiniMax-H3/FL2VA --mode dlo-dp2

-    # Rank-local mmap storage, with no DLO collective or DP wave scheduling.
-    VLLM_WORKER_MULTIPROC_METHOD=spawn \
-    VLLM_OMNI_VIDEO_SYNC_TIMEOUT=14400 \
-    CUDA_VISIBLE_DEVICES=0,1 \
-    python examples/offline_inference/minimax_h3/dlo_dp2_lifecycle.py \
-        --model /path/to/MiniMax-H3/FL2VA --mode dlo-dp2-no-allgather
-
    VLLM_WORKER_MULTIPROC_METHOD=spawn \
    VLLM_OMNI_VIDEO_SYNC_TIMEOUT=14400 \
    CUDA_VISIBLE_DEVICES=0,1 \
@@ -52,9 +45,9 @@ def parse_args() -> argparse.Namespace:
    parser.add_argument("--model", required=True, help="Path to MiniMax-H3/FL2VA")
    parser.add_argument(
        "--mode",
-        choices=("dlo-dp2", "dlo-dp2-no-allgather", "request"),
+        choices=("dlo-dp2", "request"),
        required=True,
-        help="DLO DP2 (with or without AllGather) or ordinary non-DLO TP2 request mode",
+        help="DLO AllGather DP2 or ordinary non-DLO TP2 request mode",
    )
    parser.add_argument("--steps", type=int, default=2)
    parser.add_argument("--duration", type=float, default=5.0)
@@ -81,14 +74,14 @@ def engine_kwargs(args: argparse.Namespace) -> dict[str, Any]:
        "stage_init_timeout": args.init_timeout,
        "init_timeout": args.init_timeout,
    }
-    if args.mode in {"dlo-dp2", "dlo-dp2-no-allgather"}:
+    if args.mode == "dlo-dp2":
        common.update(
            tensor_parallel_size=1,
            data_parallel_size=2,
            text_encoder_tp_size=1,
            vae_patch_parallel_size=1,
            enable_distributed_layerwise_offload=True,
-            dlo_use_allgather=args.mode == "dlo-dp2",
+            dlo_use_allgather=True,
            dlo_resident_layers=0,
        )
    else:
@@ -1,229 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""Offline speech-to-speech with NVIDIA-NemotronLabs-VoiceChat-11B.
-
-Feeds a 16 kHz user utterance (plus a spoken-style system prompt) through the
-3-stage nemotron_voicechat pipeline and writes the agent's reply as text and a
-22.05 kHz WAV.
-
-Usage:
-    python end2end.py \
-        --checkpoint /path/to/NVIDIA-NemotronLabs-VoiceChat-11B \
-        --wav /path/to/user_question_16k.wav \
-        --output-dir results/nemotron_voicechat
-
-The system prompt defaults to the NeMo reference default. The text tokenizer
-resolves from the ``nvidia/NVIDIA-Nemotron-Nano-9B-v2`` HF id (or the
-``NEMOTRON_VOICECHAT_LLM_PATH`` env var for air-gapped runs).
-
-Sizing the input WAV: the timeline is frame-locked, so the reply budget IS the
-input duration — the acoustic channel trails the text channel, and a reply
-that has not finished when the frames run out is truncated SILENTLY (no error;
-ASR of the WAV will just disagree with the emitted text). Leave generous
-trailing silence after the question (a question ending at ~4.5 s truncated in
-an 8 s WAV but completed cleanly in a 16 s one). The script warns when the
-text channel is still emitting non-PAD tokens near the last frame.
-"""
-
-from __future__ import annotations
-
-import argparse
-import json
-import os
-from pathlib import Path
-
-import numpy as np
-import soundfile as sf
-
-# Warn if the text channel is still speaking within this many final frames
-# (12 frames = ~1 s) — the frame-locked reply likely got cut off.
-TRUNCATION_GUARD_FRAMES = 12
-
-# The NeMo reference offline script's default system prompt.
-DEFAULT_SYSTEM_PROMPT = (
-    "You are an AI voice assistant developed by NVIDIA. "
-    "Your name is NVIDIA Voice Chat. "
-    "Answer in a spoken, conversational style rather than a written one. "
-    "Do not repeat the same sentence over and over again. "
-    "Start the conversation by greeting the user."
-)
-
-
-def load_wav_16k_mono(path: str) -> np.ndarray:
-    wav, sr = sf.read(path, dtype="float32")
-    if wav.ndim > 1:
-        wav = wav.mean(axis=1)
-    if sr != 16000:
-        from vllm.multimodal.audio import resample_audio_scipy
-
-        wav = resample_audio_scipy(wav, orig_sr=sr, target_sr=16000)
-    return wav.astype(np.float32)
-
-
-def main() -> None:
-    parser = argparse.ArgumentParser(description=__doc__)
-    parser.add_argument("--checkpoint", required=True, help="NVIDIA-NemotronLabs-VoiceChat-11B directory")
-    parser.add_argument("--wav", required=True, help="user speech (any sr; resampled to 16 kHz mono)")
-    parser.add_argument("--output-dir", default="results/nemotron_voicechat")
-    parser.add_argument("--system-prompt", default=DEFAULT_SYSTEM_PROMPT)
-    parser.add_argument(
-        "--stage-configs-path",
-        default=None,
-        help=(
-            "Deploy yaml override. Pass vllm_omni/deploy/nemotron_labs_voicechat_streaming.yaml "
-            "for the async-chunk streaming pipeline (stages overlap; audio starts decoding "
-            "before the thinker finishes). Default: the bundled non-streaming yaml (bit-exact "
-            "parity path)."
-        ),
-    )
-    args = parser.parse_args()
-
-    if not Path(args.wav).is_file():
-        raise FileNotFoundError(f"Input WAV not found: {args.wav}")
-    ckpt = Path(args.checkpoint)
-    raw_cfg = json.loads((ckpt / "config.json").read_text())
-    stt_cfg = raw_cfg["model"]["stt"]["model"]
-
-    from transformers import AutoTokenizer
-    from vllm.sampling_params import SamplingParams
-
-    from vllm_omni.entrypoints.omni import Omni
-    from vllm_omni.model_executor.models.nemotron_voicechat.nemotron_voicechat_thinker import (
-        compute_acoustic_frame_count,
-    )
-
-    tok_ref = os.environ.get("NEMOTRON_VOICECHAT_LLM_PATH") or stt_cfg.get(
-        "pretrained_llm", "nvidia/NVIDIA-Nemotron-Nano-9B-v2"
-    )
-    tokenizer = AutoTokenizer.from_pretrained(tok_ref, trust_remote_code=False)
-    bos_id = tokenizer.convert_tokens_to_ids(stt_cfg.get("bos_token", "<s>"))
-    eos_id = tokenizer.convert_tokens_to_ids(stt_cfg.get("eos_token", "</s>"))
-    pad_id = tokenizer.convert_tokens_to_ids(stt_cfg.get("pad_token", "<SPECIAL_12>"))
-    # NeMo encode_system_prompt: [bos] + tokens + [eos].
-    prompt_ids = [bos_id] + tokenizer.encode(args.system_prompt, add_special_tokens=False) + [eos_id]
-
-    wav = load_wav_16k_mono(args.wav)
-    n_frames = compute_acoustic_frame_count(stt_cfg, int(wav.shape[0]))
-    print(f"input: {wav.shape[0]} samples @16kHz -> {n_frames} acoustic frames; prompt {len(prompt_ids)} tokens")
-
-    inputs = {
-        # vLLM prompt = system prompt + one placeholder position (acoustic
-        # frame 0); the thinker replaces every position with fused embeddings.
-        "prompt_token_ids": prompt_ids + [pad_id],
-        "additional_information": {
-            # Plain float list: raw ndarrays do not survive the engine-core
-            # message serialization.
-            "nvc_audio": wav.astype(np.float32).tolist(),
-            "nvc_sr": 16000,
-            "nvc_prompt_token_ids": prompt_ids,
-            "nvc_expected_frames": n_frames,
-        },
-    }
-    sampling_params_list = [
-        # thinker: greedy, frame-locked generation length. NO min_tokens: the
-        # tokenizer's eos_token is <SPECIAL_12> — the frame-locked PAD/silence
-        # token — and vLLM's min_tokens masks "EOS" to -inf, which would forbid
-        # silence and force the agent to babble through the whole utterance.
-        SamplingParams(temperature=0.0, max_tokens=n_frames, ignore_eos=True, seed=0),
-        # talker: placeholder loop, stops on the stage's stop token. 16383:
-        # the 1-token placeholder prompt + max_tokens must fit the stage's
-        # max_model_len (16384).
-        SamplingParams(temperature=0.0, max_tokens=16383, stop_token_ids=[1], detokenize=False, seed=0),
-        # code2wav: single decode step.
-        SamplingParams(temperature=0.0, max_tokens=1, detokenize=False, seed=0),
-    ]
-
-    omni_kwargs: dict = {}
-    if args.stage_configs_path:
-        omni_kwargs["stage_configs_path"] = args.stage_configs_path
-    omni = Omni(model=str(ckpt), trust_remote_code=True, **omni_kwargs)
-    outputs = omni.generate(inputs, sampling_params_list)
-
-    out_dir = Path(args.output_dir)
-    out_dir.mkdir(parents=True, exist_ok=True)
-    stem = Path(args.wav).stem
-
-    agent_text = None
-    audio = None
-    sr = 22050
-    for stage_outputs in outputs:
-        # One entry per final stage (text from the thinker, audio from code2wav);
-        # mirrors the qwen2_5_omni offline example's collection loop.
-        output = getattr(stage_outputs, "request_output", stage_outputs)
-        final_type = getattr(stage_outputs, "final_output_type", None)
-        completions = getattr(output, "outputs", None) or []
-        if not completions:
-            continue
-        completion = completions[0]
-        if final_type in (None, "text"):
-            text = getattr(completion, "text", None)
-            if text:
-                agent_text = text
-            token_ids = getattr(completion, "token_ids", None)
-            if token_ids:
-                (out_dir / f"{stem}_text_tokens.json").write_text(json.dumps(list(map(int, token_ids))))
-                # Frame lock: the reply budget IS the input length. If the text
-                # channel is still mid-sentence when the frames run out, the
-                # spoken answer is silently truncated — warn instead.
-                tail = list(map(int, token_ids))[-TRUNCATION_GUARD_FRAMES:]
-                if any(t != pad_id for t in tail):
-                    print(
-                        f"WARNING: the text channel emits non-PAD tokens within the last "
-                        f"{TRUNCATION_GUARD_FRAMES} frames ({TRUNCATION_GUARD_FRAMES * 80} ms) — the reply "
-                        "likely ran out of acoustic frames and is cut off mid-sentence. "
-                        "Pad the input WAV with more trailing silence and rerun."
-                    )
-            # Best-effort: dump the function-channel debug timeline if the
-            # engine surfaced the request info on this output.
-            info = getattr(output, "additional_information", None)
-            func_timeline = info.get("nvc_function_tokens") if isinstance(info, dict) else None
-            if func_timeline is not None:
-                values = func_timeline.tolist() if hasattr(func_timeline, "tolist") else list(func_timeline)
-                (out_dir / f"{stem}_function_tokens.json").write_text(json.dumps([int(v) for v in values]))
-                print(f"function-channel timeline: {len(values)} entries")
-        if final_type in (None, "audio"):
-            # multimodal_output is a MultimodalPayload (Mapping) — duck-type it.
-            mm = getattr(completion, "multimodal_output", None)
-            get = getattr(mm, "get", None)
-            if not callable(get):
-                continue
-            candidate = None
-            for key in ("audio", "model_outputs", "wav", "waveform"):
-                value = get(key)
-                if value is not None:
-                    candidate = value
-                    break
-            if isinstance(candidate, list):
-                candidate = candidate[0] if candidate else None
-            if candidate is not None:
-                arr = np.asarray(
-                    candidate.float().cpu() if hasattr(candidate, "cpu") else candidate, dtype=np.float32
-                ).reshape(-1)
-                if arr.size > 0:
-                    audio = arr
-                sr_val = get("sr")
-                if sr_val is not None:
-                    sr_item = sr_val[0] if isinstance(sr_val, list) else sr_val
-                    sr = int(sr_item.item() if hasattr(sr_item, "item") else sr_item)
-
-    if agent_text is not None:
-        # The frame-locked text channel emits PAD tokens on silent frames;
-        # strip them (NeMo's post-inference detokenizer does the same).
-        pad_token = stt_cfg.get("pad_token", "<SPECIAL_12>")
-        agent_text = " ".join(agent_text.replace(pad_token, " ").split())
-        text_path = out_dir / f"{stem}_output.txt"
-        text_path.write_text(agent_text + "\n")
-        print(f"agent text -> {text_path}\n{agent_text}")
-    else:
-        print("WARNING: no agent text in outputs")
-
-    if audio is not None and audio.size > 0:
-        wav_path = out_dir / f"{stem}_output.wav"
-        sf.write(wav_path, audio, sr)
-        print(f"agent audio -> {wav_path} ({audio.size / sr:.2f}s @ {sr} Hz)")
-    else:
-        raise RuntimeError("No audio produced by the code2wav stage.")
-
-
-if __name__ == "__main__":
-    main()
@@ -392,7 +392,7 @@ def main(args):
    total_requests = len(prompts)
    processed_count = 0
    for stage_outputs in omni_generator:
-        output = stage_outputs
+        output = stage_outputs.request_output
        if stage_outputs.final_output_type == "text":
            request_id = output.request_id
            text_output = output.outputs[0].text
@@ -367,7 +367,7 @@ def main(args):
    print(f"query type: {args.query_type}")

    for stage_outputs in omni_generator:
-        output = stage_outputs
+        output = stage_outputs.request_output
        if stage_outputs.final_output_type == "text":
            request_id = output.request_id
            text_output = output.outputs[0].text
@@ -484,10 +484,10 @@ def parse_args():
        help="Path to a .txt file with one prompt per line (preferred).",
    )
    parser.add_argument(
-        "--deploy-config",
+        "--stage-configs-path",
        type=str,
        default=None,
-        help="Path to a deploy config YAML.",
+        help="Path to a stage configs file.",
    )
    parser.add_argument(
        "--video-path",
@@ -237,7 +237,7 @@ async def run_single_request(
            sampling_params_list=sampling_params_list,
            output_modalities=output_modalities,
        ):
-            output = omni_output
+            output = omni_output.request_output
            if omni_output.final_output_type == "text":
                if stage_0_first_output_ts is None:
                    stage_0_first_output_ts = time.perf_counter()
@@ -26,6 +26,7 @@ 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.outputs import OmniRequestOutput
 from vllm_omni.platforms import current_omni_platform


@@ -351,8 +352,8 @@ def main():
        else:
            print("[Profiler] No valid profiling data returned.")

-    # omni.generate() returns a list for sync calls; unwrap single-result list.
-    output = result[0] if isinstance(result, list) else result
+    # Extract output from result
+    output = OmniRequestOutput.unwrap_result(result)

    if not output.images:
        raise ValueError("No video frames found in OmniRequestOutput.")
@@ -232,8 +232,15 @@ def main(args):
        raise ValueError(f"Unknown query type: {args.query_type}")

    # Initialize vLLM-Omni with Step-Audio2
-    # Resolve an explicit deploy config or use the bundled task default.
-    if args.deploy_config:
+    # Resolve deploy config path. Keep explicit stage_configs_path for legacy
+    # custom configs, but use bundled deploy configs by default.
+    deploy_config_path = None
+    stage_config_path = None
+    if args.stage_configs_path and args.deploy_config:
+        raise ValueError("--stage-configs-path and --deploy-config are mutually exclusive")
+    if args.stage_configs_path:
+        stage_config_path = args.stage_configs_path
+    elif args.deploy_config:
        deploy_config_path = args.deploy_config
    else:
        configs_dir = Path(__file__).parent.parent.parent.parent / "vllm_omni/deploy"
@@ -255,7 +262,10 @@ def main(args):
        "ray_address": args.ray_address,
        "trust_remote_code": True,
    }
-    omni_kwargs["deploy_config"] = deploy_config_path
+    if stage_config_path is not None:
+        omni_kwargs["stage_configs_path"] = stage_config_path
+    else:
+        omni_kwargs["deploy_config"] = deploy_config_path

    omni_llm = Omni(**omni_kwargs)

@@ -335,7 +345,7 @@ def main(args):
    # Process outputs from each stage
    for stage_idx, stage_outputs in enumerate(omni_outputs):
        # request_output may be a single object or a list
-        raw_output = stage_outputs
+        raw_output = stage_outputs.request_output
        outputs_list = raw_output if isinstance(raw_output, list) else [raw_output]

        if stage_outputs.final_output_type == "text":
@@ -408,6 +418,12 @@ def parse_args():
            "  - HuggingFace ID: stepfun-ai/Step-Audio-2-mini"
        ),
    )
+    parser.add_argument(
+        "--stage-configs-path",
+        type=str,
+        default=None,
+        help="Path to a legacy stage config YAML file (default: use bundled deploy config)",
+    )
    parser.add_argument(
        "--deploy-config",
        type=str,
@@ -43,7 +43,6 @@ from vllm_omni.diffusion.utils.param_utils import apply_declared_extra_args
 from vllm_omni.entrypoints.omni import Omni
 from vllm_omni.inputs.data import OmniDiffusionSamplingParams
 from vllm_omni.model_extras import get_extra_body_params, get_model_class_name
-from vllm_omni.outputs import OmniRequestOutput
 from vllm_omni.platforms import current_omni_platform

 AUDIOX_TASKS = ("t2a", "t2m", "v2a", "v2m", "tv2a", "tv2m")
@@ -389,9 +388,9 @@ def main():
        raise ValueError("No output generated from omni.generate()")

    output = outputs[0]
-    if not isinstance(output, OmniRequestOutput) or not output:
+    if not hasattr(output, "request_output") or not output.request_output:
        raise ValueError("No request_output found in OmniRequestOutput")
-    request_output = output
+    request_output = output.request_output
    if not hasattr(request_output, "multimodal_output"):
        raise ValueError("No multimodal_output found in request_output")

@@ -57,9 +57,8 @@ python examples/offline_inference/text_to_image/text_to_image.py \
  --output lingbot_t2i.png
 ```

-The shared T2I entry point selects the `image` output modality. The LingBot
-pipeline validates that T2I requests produce one frame and do not include an
-input image.
+The LingBot prompt builder selects the `image` output modality and requests one
+frame. Passing an input image is not supported in LingBot T2I mode.

 ## Quick Start

@@ -115,7 +115,7 @@ def build_demo(args: argparse.Namespace) -> gr.Blocks:
        )
        images_outputs = []
        for output in outputs:
-            req_out = output
+            req_out = output.request_output
            if not isinstance(req_out, OmniRequestOutput) or not hasattr(req_out, "images"):
                raise ValueError("Invalid request_output structure or missing 'images' key")
            images = req_out.images
@@ -21,9 +21,7 @@ from vllm_omni.inputs.data import OmniDiffusionSamplingParams
 from vllm_omni.lora.request import LoRARequest
 from vllm_omni.lora.utils import stable_lora_int_id
 from vllm_omni.model_extras import (
-    build_text_to_image_prompt as build_model_text_to_image_prompt,
-)
-from vllm_omni.model_extras import (
+    build_text_to_image_prompt,
    get_extra_body_params,
    get_model_class_name,
    should_init_extra_args_for_non_diffusion_stages,
@@ -58,17 +56,6 @@ def parse_json_object(value: str, flag_name: str = "argument") -> dict[str, Any]
 parse_profiler_config = functools.partial(parse_json_object, flag_name="--profiler-config")


-def build_text_to_image_prompt(prompt: str, negative_prompt: str | None) -> dict[str, Any]:
-    """Build the canonical request envelope for the shared T2I example."""
-    result: dict[str, Any] = {
-        "prompt": prompt,
-        "modalities": ["image"],
-    }
-    if negative_prompt is not None:
-        result["negative_prompt"] = negative_prompt
-    return result
-
-
 def _normalize_images_for_save(images: list[Any]) -> list[Any]:
    """Convert NumPy diffusion outputs to PIL images before saving."""
    normalized = []
@@ -92,6 +79,12 @@ def parse_args() -> argparse.Namespace:
        "meituan-longcat/LongCat-Image, OvisAI/Ovis-Image, "
        "stabilityai/stable-diffusion-3.5-medium, Tongyi-MAI/Z-Image-Turbo and etc.",
    )
+    parser.add_argument(
+        "--stage-configs-path",
+        type=str,
+        default=None,
+        help="[Deprecated] Path to a legacy stage_args-format YAML. Prefer --deploy-config.",
+    )
    parser.add_argument(
        "--deploy-config",
        type=str,
@@ -469,6 +462,8 @@ def main():
        omni_kwargs["tensor_parallel_size"] = args.tensor_parallel_size
    if args.enforce_eager is not None:
        omni_kwargs["enforce_eager"] = args.enforce_eager
+    if args.stage_configs_path:
+        omni_kwargs["stage_configs_path"] = args.stage_configs_path
    if args.deploy_config:
        omni_kwargs["deploy_config"] = args.deploy_config
    if use_nextstep:
@@ -507,8 +502,8 @@ def main():
    print(f"  Image size: {args.width}x{args.height}")
    if args.lora_path:
        print(f"  LoRA: scale={args.lora_scale}")
-    if args.deploy_config:
-        print(f"  deploy-config: {args.deploy_config}")
+    if args.stage_configs_path:
+        print(f"  stage-configs-path: {args.stage_configs_path}")
    print(f"{'=' * 60}\n")

    # Build LoRA request when --lora-path is set
@@ -528,12 +523,9 @@ def main():
    generation_start = time.perf_counter()

    prompt_dict = build_text_to_image_prompt(
+        model_class_name=model_class_name,
        prompt=args.prompt,
        negative_prompt=args.negative_prompt,
-    )
-    prompt_dict = build_model_text_to_image_prompt(
-        model_class_name=model_class_name,
-        prompt=prompt_dict,
        height=args.height,
        width=args.width,
    )
@@ -648,7 +640,7 @@ def main():
        images = getattr(output, "images", None)
        if images:
            break
-        req_out = output
+        req_out = getattr(output, "request_output", None)
        images = getattr(req_out, "images", None) if req_out is not None else None
        if images:
            break
@@ -23,9 +23,7 @@ list of supported architectures across all modalities, see
 | OmniVoice | `k2-fsa/OmniVoice` | 2 (gen + dec) | ✓ | — | voice design, language hint | 24 kHz |
 | Qwen3-TTS | `Qwen/Qwen3-TTS-12Hz-1.7B-{CustomVoice,VoiceDesign,Base}` | 2 (talker + code2wav) | ✓ (Base) | ✓ | 3 task variants | 24 kHz |
 | VoxCPM2 | `openbmb/VoxCPM2` | single (native AR) | ✓ | ✓ (online) | continuation | 48 kHz |
-| dots.tts | `rednote-hilab/dots.tts-soar` | single (native AR) | — (not wired yet) | — | — | 48 kHz |
 | IndexTTS-2 | `IndexTeam/IndexTTS-2` | 2 (AR talker + S2Mel DiT + BigVGAN) | ✓ (required) | — | emotion control (`--emo-audio`, `--emo-text`, `--emo-vector`) | 22.05 kHz |
-| IndexTTS-2.5 | native `checkpoints/` bundle | 2 (AR talker + EnhancedCodec + S2Mel DiT + BigVGAN) | ✓ (required) | — | multilingual (`--lang`) + emotion control | 22.05 kHz |
 | Voxtral TTS | `mistralai/Voxtral-4B-TTS-2603` | varies | ✓ | ✓ | voice presets | 24 kHz |

 ## Common Quick Start
@@ -373,10 +371,18 @@ python examples/offline_inference/text_to_speech/qwen3_tts/end2end.py \
 Streaming requires `async_chunk: true` in the stage config.

 ### Word Timestamps
-Word-level timestamps are currently a serving-path feature: launch
-`vllm-omni serve` with `--forced-aligner` and request `word_timestamps`
-(see `examples/online_serving/text_to_speech/README.md`). An offline
-example is not provided in this release.
+Generate a WAV offline and a JSON sidecar with word-level timestamps from
+`Qwen/Qwen3-ForcedAligner-0.6B`:
+```bash
+python examples/offline_inference/text_to_speech/qwen3_tts/word_timestamps.py \
+    --model Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice \
+    --forced-aligner Qwen/Qwen3-ForcedAligner-0.6B \
+    --text "Hello world." \
+    --output-dir /tmp/qwen3_tts_timestamps
+```
+The script writes `qwen3_tts_word_timestamps.wav` and
+`qwen3_tts_word_timestamps.json`. On machines without a local CUDA toolkit,
+set `VLLM_USE_FLASHINFER_SAMPLER=0` to avoid FlashInfer sampler JIT.

 ### Batched decoding
 The Code2Wav stage supports batched decoding through the SpeechTokenizer. Pass multiple prompts via `--txt-prompts` and set `--batch-size` accordingly. To raise `max_num_seqs` on either stage, point `--stage-configs-path` at a stage configs YAML with the desired values (see `vllm_omni/model_executor/stage_configs/` for templates):
@@ -431,39 +437,11 @@ Streaming is exposed through the online OpenAI Speech API (`stream=true`). See [

 ---

-## dots.tts
+## IndexTTS-2

-Single-stage native AR TTS at 48 kHz (rednote-hilab). Pipeline: `Qwen2.5-1.5B base LM → DiT (10-step Euler flow matching) → patch_encoder AR loopback → AudioVAE (streaming decode)`. Same "vLLM-native base LM + side-path computation" pattern as VoxCPM2, with a plain Qwen2 backbone instead of MiniCPM4 and no FSQ / residual LM stage.
+2-stage TTS pipeline (GPT AR talker + S2Mel CFM DiT + BigVGAN vocoder) at 22.05 kHz. Every request requires reference audio for zero-shot voice cloning. Supports emotion conditioning via audio, text, or 8-dim vector.

 ### Quick start
-```bash
-python examples/offline_inference/text_to_speech/dots_tts/end2end.py \
-    --model rednote-hilab/dots.tts-soar \
-    --text "Hello, this is a test of dots TTS running on vLLM Omni."
-```
-
-### Voice cloning
-Not wired in this release — generation is zero-shot only. The CAM++ x-vector speaker encoder weights load, but `end2end.py` has no `--ref-audio`/`--ref-text` flags yet.
-
-### Streaming
-The AudioVAE decoder has an internal streaming path (`init_stream_state` / `stream_step` / `stream_flush`) used to avoid boundary artifacts between 160 ms patches, but it is not yet exposed through an online serving endpoint or example.
-
-### Notes
- Output: 48 kHz mono WAV.
- Deploy config: `vllm_omni/deploy/dots_tts.yaml` (auto-loaded by HF `model_type`).
- Checkpoints: `rednote-hilab/dots.tts-soar` is the validated default. `dots.tts-base` shares the same architecture but is unvalidated in this repo. `dots.tts-mf` (MeanFlow, 2-4 step) is not supported yet.
- Known limitation: no CUDA graph capture and no batched side-path computation yet, so concurrent requests do not currently scale (each request's DiT Euler steps run serially). See `recipes/rednote-hilab/dots.tts.md` for details and the roadmap.
-
---
-
-## IndexTTS-2 and IndexTTS-2.5
-
-Both versions use a 2-stage TTS pipeline and produce 22.05 kHz mono audio. Stage 0 is a GPT AR talker. IndexTTS-2 Stage 1 uses RepCodec semantic embeddings, GPT latent, S2Mel CFM/DiT, and BigVGAN. IndexTTS-2.5 replaces RepCodec with EnhancedCodec and uses the official code-only path (`use_gpt_latent=false`). Every request requires reference audio for zero-shot voice cloning.
-
-### Quick start
-
-IndexTTS-2:
-
 ```bash
 python examples/offline_inference/text_to_speech/indextts2/end2end.py \
    --model IndexTeam/IndexTTS-2 \
@@ -471,24 +449,6 @@ python examples/offline_inference/text_to_speech/indextts2/end2end.py \
    --ref-audio /path/to/reference.wav
 ```

-IndexTTS-2 uses its existing frontend and does not require the IndexTTS-2.5 text dependencies. Before running IndexTTS-2.5, install its optional frontend dependencies:
-
-```bash
-pip install 'vllm-omni[indextts2]'
-```
-
-IndexTTS-2.5 with a native repository/lab bundle whose model root contains `checkpoints/`:
-
-```bash
-python examples/offline_inference/text_to_speech/indextts2/end2end.py \
-    --model /path/to/indextts-2.5 \
-    --model-version 2.5 \
-    --lang ja \
-    --text "こんにちは、音声合成のテストです。" \
-    --ref-audio /path/to/reference.wav
-
-```
-
 ### Emotion control
 ```bash
 # Emotion from reference audio
@@ -514,13 +474,10 @@ python examples/offline_inference/text_to_speech/indextts2/end2end.py \
 ```

 ### Notes
- `--ref-audio` is **required** — neither version provides text-only synthesis.
- IndexTTS-2.5 Stage 0 (AR Talker) generates mel codes from text + reference audio using plain vLLM sampling. It does not reproduce the official default `num_beams=3` beam search. For parity comparisons, run upstream with `num_beams=1`; output quality can differ from the official beam-search result.
- IndexTTS-2.5 accepts language codes such as `zh`, `en`, `zhen` (mixed Chinese/English), `ja`, and `yue`. `Mandarin` is a vLLM-Omni convenience alias for `zh`. Japanese (`ja`) uses `fugashi` tokenization and produces audio, but it does not automatically expand numbers, dates, or percentages; write those inputs as readable Japanese text before calling the model. Use `--no-text-normalization` only when the input is already normalized.
- A seed controls Stage 0 AR sampling and per-request CFM noise. Different concurrent batch compositions do not guarantee a bit-identical waveform.
- Stage 1 (semantic codec + S2Mel + BigVGAN): CFM/DiT converts the complete semantic-code sequence to waveform.
- Deploy configs: `vllm_omni/deploy/indextts2.yaml` and `indextts2_5.yaml`.
- `async_chunk=false` is intentional for IndexTTS-2.5 correctness: S2Mel consumes the completed semantic sequence.
+- `--ref-audio` is **required** — IndexTTS-2 does not support text-only synthesis.
+- Stage 0 (AR Talker): GPT-2 generates mel codes from text + reference audio.
+- Stage 1 (S2Mel + BigVGAN): CFM DiT converts mel codes to waveform at 22.05 kHz.
+- Deploy config: `vllm_omni/deploy/indextts2.yaml`. Stage 1 runs with `enforce_eager: true` (DiT has dynamic shapes).

 ---

@@ -168,7 +168,7 @@ def run_e2e():
    print(f"Received {len(outputs)} outputs.")
    for i, output in enumerate(outputs):
        try:
-            ro = output
+            ro = output.request_output
            if ro is None:
                print("No request_output found.")
                continue
@@ -1,125 +0,0 @@
-"""Offline dots.tts inference example (single-stage native AR pipeline).
-
-Uses the single-stage native AR config (vllm_omni/deploy/dots_tts.yaml).
-Reference checkpoint: rednote-hilab/dots.tts-soar.
-"""
-
-from __future__ import annotations
-
-import os
-import time
-from pathlib import Path
-
-import soundfile as sf
-import torch
-
-from vllm_omni import Omni
-from vllm_omni.utils.tracking_parser import TrackingArgumentParser
-
-REPO_ROOT = Path(__file__).resolve().parents[4]
-SAMPLE_RATE = 48_000
-
-
-def parse_args():
-    parser = TrackingArgumentParser(description="Offline dots.tts native AR inference")
-    parser.add_argument(
-        "--model",
-        type=str,
-        default="rednote-hilab/dots.tts-soar",
-        help="dots.tts model path or HuggingFace repo ID.",
-    )
-    parser.add_argument(
-        "--text",
-        type=str,
-        default="Hello, this is a test of dots TTS running on vLLM Omni.",
-        help="Text to synthesize.",
-    )
-    parser.add_argument(
-        "--output-dir",
-        type=str,
-        default="output_audio",
-        help="Directory for output WAV files.",
-    )
-    parser.add_argument(
-        "--deploy-config",
-        type=str,
-        default=None,
-        help="Override the deploy config path. If unset, auto-loads "
-        "vllm_omni/deploy/dots_tts.yaml based on the HF model_type.",
-    )
-    return parser.parse_args()
-
-
-def extract_audio(multimodal_output: dict) -> torch.Tensor:
-    """Extract the final complete audio tensor from multimodal output.
-
-    The output processor concatenates per-step delta tensors under
-    ``model_outputs``.  Falls back to ``audio`` for backwards compat.
-    """
-    audio = multimodal_output.get("model_outputs")
-    if audio is None:
-        audio = multimodal_output.get("audio")
-    if audio is None:
-        raise ValueError(f"No audio key in multimodal_output: {list(multimodal_output.keys())}")
-
-    if isinstance(audio, list):
-        valid = [torch.as_tensor(a).float().cpu().reshape(-1) for a in audio if a is not None]
-        if not valid:
-            raise ValueError("Audio list is empty or all elements are None.")
-        return torch.cat(valid, dim=0) if len(valid) > 1 else valid[0]
-
-    return torch.as_tensor(audio).float().cpu().reshape(-1)
-
-
-def main():
-    args = parse_args()
-
-    output_dir = Path(args.output_dir)
-    output_dir.mkdir(parents=True, exist_ok=True)
-
-    engine = Omni(
-        model=args.model,
-        deploy_config=args.deploy_config,
-    )
-
-    from transformers import AutoTokenizer
-
-    from vllm_omni.model_executor.models.dots_tts.dots_tts_prompt import (
-        build_dots_tts_prompt,
-    )
-
-    tokenizer = AutoTokenizer.from_pretrained(args.model, trust_remote_code=True)
-
-    prompt = build_dots_tts_prompt(
-        tokenizer=tokenizer,
-        text=args.text,
-    )
-
-    print(f"Model       : {args.model}")
-    print(f"Text        : {args.text}")
-    print(f"Output dir  : {output_dir}")
-    print(f"Prompt len  : {len(prompt['prompt_token_ids'])} tokens")
-
-    t_start = time.perf_counter()
-    outputs = engine.generate([prompt])
-    elapsed = time.perf_counter() - t_start
-
-    request_output = outputs[0]
-    mm = request_output.outputs[0].multimodal_output
-    audio = extract_audio(mm)
-
-    duration = audio.numel() / SAMPLE_RATE
-    rtf = elapsed / duration if duration > 0 else float("inf")
-
-    output_path = output_dir / "output.wav"
-    sf.write(str(output_path), audio.numpy(), SAMPLE_RATE, format="WAV")
-
-    print(f"Saved       : {output_path}")
-    print(f"Duration    : {duration:.2f}s")
-    print(f"Inference   : {elapsed:.2f}s")
-    print(f"RTF         : {rtf:.3f}")
-
-
-if __name__ == "__main__":
-    os.environ["VLLM_WORKER_MULTIPROC_METHOD"] = "spawn"
-    main()
@@ -38,7 +38,7 @@ from vllm_omni.utils.tracking_parser import TrackingArgumentParser
 logger = logging.getLogger(__name__)

 DEFAULT_MODEL = "fishaudio/s2-pro"
-DEFAULT_DEPLOY_CONFIG = os.path.join(
+DEFAULT_STAGE_CONFIG = os.path.join(
    os.path.dirname(__file__),
    "..",
    "..",
@@ -152,7 +152,10 @@ def _build_omni_kwargs(args, model_name: str) -> dict:
        log_stats=args.log_stats,
        stage_init_timeout=args.stage_init_timeout,
    )
-    omni_kwargs["deploy_config"] = args.deploy_config or DEFAULT_DEPLOY_CONFIG
+    if args.deploy_config:
+        omni_kwargs["deploy_config"] = args.deploy_config
+    else:
+        omni_kwargs["stage_configs_path"] = args.stage_configs_path or DEFAULT_STAGE_CONFIG
    return omni_kwargs


@@ -176,7 +179,7 @@ def main(args):

    t_start = time.perf_counter()
    for stage_outputs in omni.generate(inputs):
-        request_output = stage_outputs
+        request_output = stage_outputs.request_output
        if request_output is None or not request_output.outputs:
            continue
        _save_wav(
@@ -212,7 +215,7 @@ async def main_streaming(args):
    sample_rate = None

    async for stage_output in omni.generate(prompt, request_id=request_id):
-        mm = stage_output.outputs[0].multimodal_output
+        mm = stage_output.request_output.outputs[0].multimodal_output
        if not stage_output.finished:
            t_now = time.perf_counter()
            sample_rate, n = _collect_audio_chunk(mm, all_audio_chunks, sample_rate)
@@ -275,6 +278,12 @@ def parse_args():
        default=None,
        help="Deploy config YAML path (under vllm_omni/deploy/).",
    )
+    parser.add_argument(
+        "--stage-configs-path",
+        type=str,
+        default=None,
+        help="Path to stage configs YAML (legacy, prefer --deploy-config).",
+    )
    parser.add_argument(
        "--output-dir",
        default="output_audio",
@@ -145,7 +145,7 @@ def main(args):

    omni = Omni(
        model=args.model,
-        deploy_config=deploy_config_path,
+        stage_configs_path=deploy_config_path,
        log_stats=args.log_stats,
        stage_init_timeout=args.stage_init_timeout,
    )
@@ -1,16 +1,13 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""Offline inference example for IndexTTS 2.0 and 2.5 via vLLM-Omni.
+"""Offline inference example for IndexTTS2 via vLLM-Omni.

-Two-stage pipeline: GPT AR (Stage 0) → semantic codec + S2Mel + BigVGAN
-(Stage 1).
+Two-stage pipeline: GPT AR (Stage 0) → S2Mel + BigVGAN (Stage 1).
 Output is 22050 Hz mono WAV.

 Usage:
  python end2end.py \
-    --model /path/to/indextts-2.5 \
-    --model-version 2.5 \
-    --lang zh \
+    --model /path/to/IndexTeam/IndexTTS-2 \
    --text "你好，这是一个语音合成测试。" \
    --ref-audio /path/to/ref.wav

@@ -50,10 +47,6 @@ def build_request(
    emo_alpha: float | None = None,
    use_emo_text: bool = False,
    use_random: bool = False,
-    model_type: str = "indextts2",
-    lang: str = "zh",
-    text_normalization: bool = True,
-    tokenizer_file: str | None = None,
 ) -> dict:
    additional: dict = {"text": [text]}
    if ref_audio_path:
@@ -70,44 +63,12 @@ def build_request(
        additional["use_emo_text"] = [True]
    if use_random:
        additional["use_random"] = [True]
-    if model_type == "indextts2_5":
-        additional["lang"] = [lang]
-        additional["text_normalization"] = [text_normalization]
-    prompt_kwargs = {
-        "model_type": model_type,
-        "lang": lang,
-        "text_normalization": text_normalization,
-    }
-    if tokenizer_file is not None:
-        prompt_kwargs["tokenizer_file"] = tokenizer_file
    return {
-        "prompt_token_ids": build_indextts2_prefill_prompt_ids(
-            model,
-            text,
-            **prompt_kwargs,
-        ),
+        "prompt_token_ids": build_indextts2_prefill_prompt_ids(model, text),
        "additional_information": additional,
    }


-def _get_stage0_tokenizer_file(omni: Omni) -> str | None:
-    """Read the effective tokenizer override from the resolved Stage 0 config."""
-    stage_configs = omni.stage_configs
-    if not stage_configs:
-        return None
-    engine_args = getattr(stage_configs[0], "engine_args", None)
-    if engine_args is None:
-        return None
-    hf_overrides = getattr(engine_args, "hf_overrides", None)
-    if hf_overrides is None and isinstance(engine_args, dict):
-        hf_overrides = engine_args.get("hf_overrides")
-    if hf_overrides is None:
-        return None
-    if hasattr(hf_overrides, "get"):
-        return hf_overrides.get("tokenizer_file")
-    return getattr(hf_overrides, "tokenizer_file", None)
-
-
 def save_audio(waveform: torch.Tensor, path: str, sample_rate: int = 22050) -> None:
    audio_np = waveform.float().numpy()
    sf.write(path, audio_np, sample_rate)
@@ -134,14 +95,9 @@ def extract_audio(mm: dict) -> tuple[torch.Tensor | None, int]:


 def main(args) -> None:
-    model_type = "indextts2_5" if args.model_version == "2.5" else "indextts2"
-    deploy_config = args.deploy_config
-    if deploy_config is None and args.model_version == "2.5":
-        repo_root = Path(__file__).resolve().parents[4]
-        deploy_config = str(repo_root / "vllm_omni" / "deploy" / "indextts2_5.yaml")
    omni = Omni(
        model=args.model,
-        deploy_config=deploy_config,
+        deploy_config=args.deploy_config,
        stage_init_timeout=args.stage_init_timeout,
    )

@@ -170,7 +126,7 @@ def main(args) -> None:
    print(f"Synthesizing: {args.text!r}")
    if args.ref_audio:
        print(f"  ref_audio: {args.ref_audio}")
-    request_kwargs = dict(
+    inputs = build_request(
        model=args.model,
        text=args.text,
        ref_audio_path=args.ref_audio,
@@ -180,17 +136,7 @@ def main(args) -> None:
        emo_alpha=args.emo_alpha,
        use_emo_text=args.use_emo_text,
        use_random=args.use_random,
-        model_type=model_type,
-        lang=args.lang,
-        text_normalization=args.text_normalization,
    )
-    if model_type == "indextts2_5":
-        tokenizer_file = _get_stage0_tokenizer_file(omni)
-        if tokenizer_file is not None:
-            request_kwargs["tokenizer_file"] = tokenizer_file
-    inputs = build_request(**request_kwargs)
-    if model_type == "indextts2_5":
-        inputs["additional_information"]["duration_factor"] = [1.0 / getattr(args, "speed", 1.0)]

    for i, omni_out in enumerate(omni.generate(inputs, sampling_params_list=sampling_params)):
        mm = omni_out.multimodal_output
@@ -208,37 +154,8 @@ def main(args) -> None:


 def parse_args():
-    parser = FlexibleArgumentParser(description="IndexTTS 2.0/2.5 offline inference")
-    parser.add_argument(
-        "--model",
-        required=True,
-        help="HF model path or native bundle containing checkpoints/.",
-    )
-    parser.add_argument(
-        "--model-version",
-        choices=("2.0", "2.5"),
-        default="2.0",
-    )
-    parser.add_argument(
-        "--lang",
-        default="zh",
-        help=(
-            "IndexTTS 2.5 language code, for example zh/en/zhen/ja/yue; "
-            "zhen is mixed Chinese/English and Mandarin is a vLLM-Omni alias for zh."
-        ),
-    )
-    parser.add_argument(
-        "--no-text-normalization",
-        action="store_false",
-        dest="text_normalization",
-        help="Disable IndexTTS 2.5 text normalization.",
-    )
-    parser.add_argument(
-        "--speed",
-        type=float,
-        default=1.0,
-        help="IndexTTS 2.5 native synthesis speed in [0.5, 2.0]; 2.0 is faster.",
-    )
+    parser = FlexibleArgumentParser(description="IndexTTS2 offline inference")
+    parser.add_argument("--model", required=True, help="HF model path for IndexTTS2.")
    parser.add_argument("--text", default="你好，这是IndexTTS2语音合成测试。")
    parser.add_argument("--ref-audio", required=True, help="Reference audio for voice cloning.")
    parser.add_argument("--emo-audio", default=None, help="Emotion reference audio.")
@@ -253,15 +170,7 @@ def parse_args():
    parser.add_argument("--emo-alpha", type=float, default=None, help="Emotion weight in [0, 1].")
    parser.add_argument("--use-emo-text", action="store_true", help="Infer emotion vector from emo-text or text.")
    parser.add_argument("--use-random", action="store_true", help="Use random emotion prototypes.")
-    parser.add_argument(
-        "--seed",
-        type=int,
-        default=None,
-        help=(
-            "Seed Stage 0 AR sampling and per-request CFM noise; changing the "
-            "concurrent batch composition may still change the final waveform."
-        ),
-    )
+    parser.add_argument("--seed", type=int, default=None)
    parser.add_argument(
        "--output-dir",
        default=os.path.join(
@@ -271,10 +180,7 @@ def parse_args():
    )
    parser.add_argument("--deploy-config", default=None)
    parser.add_argument("--stage-init-timeout", type=int, default=600)
-    args = parser.parse_args()
-    if args.model_version == "2.5" and not 0.5 <= args.speed <= 2.0:
-        parser.error("IndexTTS 2.5 --speed must be between 0.5 and 2.0")
-    return args
+    return parser.parse_args()


 if __name__ == "__main__":
@@ -137,7 +137,7 @@ def main(args) -> None:
    params_list = sampling_params

    for stage_outputs in omni.generate(inputs, params_list):
-        for i, req_output in enumerate(stage_outputs):
+        for i, req_output in enumerate(stage_outputs.request_output):
            for j, out in enumerate(req_output.outputs):
                mm = out.multimodal_output
                if mm is None:
@@ -135,7 +135,7 @@ def run_e2e():
    print(f"Received {len(outputs)} outputs.")
    for i, output in enumerate(outputs):
        try:
-            ro = output
+            ro = output.request_output
            if ro is None:
                print("No request_output found.")
                continue
@@ -391,7 +391,7 @@ def main(args):
    for batch_start in range(0, len(inputs), batch_size):
        batch = inputs[batch_start : batch_start + batch_size]
        for stage_outputs in omni.generate(batch):
-            output = stage_outputs
+            output = stage_outputs.request_output
            _save_wav(output_dir, output.request_id, output.outputs[0].multimodal_output)


@@ -412,7 +412,7 @@ async def main_streaming(args):
        t_prev = t_start
        chunk_idx = 0
        async for stage_output in omni.generate(prompt, request_id=request_id):
-            mm = stage_output.outputs[0].multimodal_output
+            mm = stage_output.request_output.outputs[0].multimodal_output
            if not stage_output.finished:
                t_now = time.perf_counter()
                audio = mm.get("audio")
@@ -490,10 +490,10 @@ def parse_args():
        help="Path to a .txt file with one prompt per line (preferred).",
    )
    parser.add_argument(
-        "--deploy-config",
+        "--stage-configs-path",
        type=str,
        default=None,
-        help="Path to a deploy config YAML.",
+        help="Path to a stage configs file.",
    )
    parser.add_argument(
        "--audio-path",
@@ -116,7 +116,7 @@ def main(args: Any) -> None:
    prompt = _build_custom_voice_input(args)
    final_output = None
    for stage_outputs in omni.generate([prompt]):
-        final_output = stage_outputs
+        final_output = stage_outputs.request_output
    if final_output is None:
        raise RuntimeError("Qwen3-TTS did not produce an output.")

@@ -38,7 +38,7 @@ def _stage_durations_from_output(outputs) -> dict[str, float]:
    durations = getattr(omni_out, "stage_durations", None) or {}
    if durations:
        return dict(durations)
-    ro = omni_out
+    ro = getattr(omni_out, "request_output", None)
    if ro is not None:
        return dict(getattr(ro, "stage_durations", None) or {})
    return {}
@@ -99,7 +99,7 @@ def extract_audio(outputs) -> tuple[np.ndarray, int]:
    out = outputs[0]
    mm = getattr(out, "multimodal_output", None)
    if mm is None:
-        ro = out
+        ro = getattr(out, "request_output", None)
        if ro is not None:
            mm = getattr(ro, "multimodal_output", None)
            if mm is None and ro.outputs:
@@ -8,7 +8,6 @@ A unified script for text-to-video generation. Supports multiple models with mod
 |---|---|---|---|---|---|
 | `Wan-AI/Wan2.1-VACE-1.3B-diffusers` | 480x832 | 81 | 30 | 5.0 | ~20 GiB (RTX 5090, VAE tiling) |
 | `Wan-AI/Wan2.2-T2V-A14B-Diffusers` | 720x1280 | 81 | 40 | 4.0 | ~60 GiB |
-| `robbyant/lingbot-video-dense-1.3b` / `robbyant/lingbot-video-moe-30b-a3b` | 192x320 | 9 | 2 | 3.0 | ~68 GiB (MoE smoke) |
 | `Lightricks/LTX-2` | 512x768 | 121 | 40 | video 3.0 / audio 7.0 | Model-dependent |
 | `hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v` | 480x832 | 121 | 50 | 6.0 | 1×A100 80GB |
 | `hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v` | 720x1280 | 121 | 50 | 6.0 | FP8 + VAE tiling required |
@@ -58,31 +57,18 @@ python text_to_video.py \

 ### LingBot-Video

-The shared runner recognizes the official dense and MoE checkpoint IDs, selects
-`LingBotVideoPipeline`, and builds the canonical video request envelope:
+The dedicated LingBot example explicitly selects the video output modality:

 ```bash
-python text_to_video.py \
+python examples/offline_inference/text_to_video/text_to_video_lingbot.py \
  --model robbyant/lingbot-video-dense-1.3b \
-  --prompt "a robotic arm picks up a red block" \
  --height 192 --width 320 --num-frames 9 --num-inference-steps 2 \
-  --guidance-scale 3.0 --flow-shift 3.0 --fps 24 \
  --output lingbot_t2v.mp4
 ```

 Use the same command with `robbyant/lingbot-video-moe-30b-a3b` for the MoE
 checkpoint. Requested frame counts are rounded upward to the causal
-VAE's `4n+1` grid. For a local checkpoint path whose name does not contain
-`lingbot`, pass `--model-class-name LingBotVideoPipeline`.
-
-LingBot-only options such as `batch_cfg` and `output_type` use the shared
-model-extra channel, for example:
-
-```bash
-python text_to_video.py \
-  --model robbyant/lingbot-video-dense-1.3b \
-  --extra-body '{"batch_cfg": true, "output_type": "np"}'
-```
+VAE's `4n+1` grid.

 ### LTX-2

@@ -16,7 +16,7 @@ from vllm_omni.entrypoints.omni import Omni
 from vllm_omni.inputs.data import OmniDiffusionSamplingParams
 from vllm_omni.lora.request import LoRARequest
 from vllm_omni.lora.utils import stable_lora_int_id
-from vllm_omni.model_extras import get_extra_body_params, get_model_class_name, get_output_tensor_range
+from vllm_omni.model_extras import get_extra_body_params, get_model_class_name
 from vllm_omni.outputs import OmniRequestOutput
 from vllm_omni.platforms import current_omni_platform

@@ -80,17 +80,6 @@ _MODEL_PRESETS = {
        "fps": 16,
        "output": "helios_output.mp4",
    },
-    "lingbot": {
-        "model_class_name": "LingBotVideoPipeline",
-        "height": 192,
-        "width": 320,
-        "num_frames": 9,
-        "num_inference_steps": 2,
-        "guidance_scale": 3.0,
-        "fps": 24,
-        "flow_shift": 3.0,
-        "output": "lingbot_video_output.mp4",
-    },
    "ltx2": {
        "height": 512,
        "width": 768,
@@ -121,51 +110,27 @@ _MODEL_PRESETS = {
 def _detect_preset(model: str, model_class_name: str | None = None) -> dict:
    model_lower = model.lower()
    class_lower = (model_class_name or "").lower()
-    if "lingbot" in model_lower or "lingbotvideo" in class_lower:
-        return _MODEL_PRESETS["lingbot"]
-    if "ltx" in class_lower or "ltx" in model_lower:
-        if "distilled" in class_lower or "distilled" in model_lower:
-            return _MODEL_PRESETS["ltx2_distilled"]
-        if "ltx23" in class_lower or "ltx-2.3" in model_lower or "ltx_2.3" in model_lower:
-            return _MODEL_PRESETS["ltx23"]
+    if "distilled" in class_lower or "distilled" in model_lower:
+        return _MODEL_PRESETS["ltx2_distilled"]
+    if "ltx23" in class_lower or "ltx-2.3" in model_lower or "ltx_2.3" in model_lower:
+        return _MODEL_PRESETS["ltx23"]
+    if "ltx2" in class_lower or "ltx-2" in model_lower or "ltx_2" in model_lower:
        return _MODEL_PRESETS["ltx2"]
-    if "vace" in model_lower or "vace" in class_lower:
+    if "vace" in model_lower:
        return _MODEL_PRESETS["vace"]
    # Edge must be matched before the generic cosmos branch (its "cosmos" substring would
    # otherwise pick up the Nano/Super 720p / gs 6.0 / flow_shift 10.0 preset).
    if ("cosmos" in model_lower or "cosmos" in class_lower) and ("edge" in model_lower or "edge" in class_lower):
        return _MODEL_PRESETS["cosmos3_edge"]
-    if "cosmos" in model_lower or "cosmos" in class_lower:
+    if "cosmos" in model_lower:
        return _MODEL_PRESETS["cosmos"]
-    if "hunyuan" in model_lower or "hunyuan" in class_lower:
+    if "hunyuan" in model_lower:
        return _MODEL_PRESETS["hunyuan"]
-    if "helios" in model_lower or "helios" in class_lower:
+    if "helios" in model_lower:
        return _MODEL_PRESETS["helios"]
    return _MODEL_PRESETS["wan"]


-def build_text_to_video_prompt(prompt: str, negative_prompt: str | None) -> dict[str, Any]:
-    """Build the canonical request envelope for the shared T2V example."""
-    result: dict[str, Any] = {
-        "prompt": prompt,
-        "modalities": ["video"],
-    }
-    if negative_prompt is not None:
-        result["negative_prompt"] = negative_prompt
-    return result
-
-
-def _normalize_float_tensor(tensor: torch.Tensor, source_range: str) -> torch.Tensor:
-    """Normalize decoded frames according to the pipeline's output contract."""
-    if not tensor.is_floating_point():
-        return tensor
-    if source_range == "negative_one_to_one":
-        return tensor.clamp(-1, 1) * 0.5 + 0.5
-    if source_range == "zero_to_one":
-        return tensor.clamp(0, 1)
-    raise ValueError(f"Unsupported floating-point tensor range: {source_range!r}")
-
-
 def parse_profiler_config(value: str) -> dict[str, Any]:
    try:
        config = json.loads(value)
@@ -195,7 +160,7 @@ def parse_extra_body(value: str) -> dict[str, Any]:
 def parse_args() -> argparse.Namespace:
    parser = argparse.ArgumentParser(
        description="Generate a video from a text prompt. "
-        "Supports Wan2.2, HunyuanVideo-1.5, Helios, LingBot-Video, and other text-to-video models."
+        "Supports Wan2.2, HunyuanVideo-1.5, Helios, and other text-to-video models."
    )
    parser.add_argument(
        "--model",
@@ -425,7 +390,7 @@ def _extract_peak_memory_mb(result: Any) -> float:
        return 0.0
    val = getattr(result, "peak_memory_mb", 0.0)
    if not val:
-        inner = result
+        inner = getattr(result, "request_output", None)
        if isinstance(inner, list):
            inner = inner[0] if inner else None
        val = getattr(inner, "peak_memory_mb", 0.0)
@@ -443,7 +408,6 @@ def main():
    for key, default_val in preset.items():
        if getattr(args, key.replace("-", "_"), None) is None:
            setattr(args, key.replace("-", "_"), default_val)
-    model_class_name = args.model_class_name

    generator = torch.Generator(device=current_omni_platform.device_type).manual_seed(args.seed)
    # Cache-dit config (Wan2.2 only)
@@ -516,7 +480,6 @@ def main():
    omni = Omni(**omni_kwargs)
    model_class_name = get_model_class_name(omni) or model_class_name
    declared_extra_body_params = get_extra_body_params(model_class_name)
-    output_tensor_range = get_output_tensor_range(model_class_name)

    if profiler_enabled:
        print("[Profiler] Starting profiling...")
@@ -550,11 +513,12 @@ def main():
            lora_path=lora_path,
        )

-    negative_prompt = args.negative_prompt
-    if negative_prompt is None and all(preset is not _MODEL_PRESETS[name] for name in ("lingbot", "ltx2", "ltx23")):
+    prompt_dict = {"prompt": args.prompt}
+    if args.negative_prompt is not None:
+        prompt_dict["negative_prompt"] = args.negative_prompt
+    elif preset not in (_MODEL_PRESETS["ltx2"], _MODEL_PRESETS["ltx23"]):
        # Preserve the historical empty-prompt behavior for non-LTX examples.
-        negative_prompt = ""
-    prompt_dict = build_text_to_video_prompt(args.prompt, negative_prompt)
+        prompt_dict["negative_prompt"] = ""

    extra_args = {}
    if lora_request:
@@ -611,8 +575,8 @@ def main():
        if frames.multimodal_output and "audio" in frames.multimodal_output:
            audio = frames.multimodal_output["audio"]
            audio_sample_rate = frames.multimodal_output.get("audio_sample_rate", audio_sample_rate)
-        if frames.is_pipeline_output and frames is not None:
-            inner_output = frames
+        if frames.is_pipeline_output and frames.request_output is not None:
+            inner_output = frames.request_output
            if isinstance(inner_output, OmniRequestOutput):
                if inner_output.multimodal_output and "audio" in inner_output.multimodal_output:
                    audio = inner_output.multimodal_output["audio"]
@@ -666,7 +630,8 @@ def main():
                frame_tensor = frame_tensor[0]
            if frame_tensor.dim() == 3 and frame_tensor.shape[0] in (3, 4):
                frame_tensor = frame_tensor.permute(1, 2, 0)
-            frame_tensor = _normalize_float_tensor(frame_tensor, output_tensor_range)
+            if frame_tensor.is_floating_point():
+                frame_tensor = frame_tensor.clamp(-1, 1) * 0.5 + 0.5
            return frame_tensor.float().numpy()
        if isinstance(frame, np.ndarray):
            frame_array = frame
@@ -718,7 +683,8 @@ def main():
                video_tensor = video_tensor[0]
        elif video_tensor.dim() == 4 and video_tensor.shape[0] in (3, 4):
            video_tensor = video_tensor.permute(1, 2, 3, 0)
-        video_tensor = _normalize_float_tensor(video_tensor, output_tensor_range)
+        if video_tensor.is_floating_point():
+            video_tensor = video_tensor.clamp(-1, 1) * 0.5 + 0.5
        video_array = video_tensor.float().numpy()
    elif isinstance(frames, np.ndarray):
        video_array = frames
@@ -0,0 +1,131 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+
+from __future__ import annotations
+
+import argparse
+import time
+from pathlib import Path
+from typing import Any
+
+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.outputs import OmniRequestOutput
+from vllm_omni.platforms import current_omni_platform
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="Generate text-to-video with a LingBot-Video checkpoint.")
+    parser.add_argument("--model", default="robbyant/lingbot-video-dense-1.3b")
+    parser.add_argument("--prompt", default="a robotic arm picks up a red block")
+    parser.add_argument("--negative-prompt", default=None)
+    parser.add_argument("--output", default="lingbot_video_output.mp4")
+    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("--num-inference-steps", type=int, default=2)
+    parser.add_argument("--guidance-scale", type=float, default=3.0)
+    parser.add_argument("--flow-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("--output-type", choices=["pt", "np"], default="pt")
+    parser.add_argument("--batch-cfg", action="store_true")
+    return parser.parse_args()
+
+
+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 _extract_video(output: Any) -> Any:
+    if isinstance(output, list):
+        output = output[0] if output else None
+    if isinstance(output, OmniRequestOutput):
+        if output.final_output_type != "image":
+            raise ValueError(f"Unexpected output type {output.final_output_type!r}.")
+        if not output.images:
+            raise ValueError("No video frames found in OmniRequestOutput.")
+        return output.images[0] if len(output.images) == 1 else output.images
+    return output
+
+
+def _to_frame_array(video: Any) -> np.ndarray:
+    if isinstance(video, torch.Tensor):
+        tensor = video.detach().cpu()
+        if tensor.dim() == 5:
+            tensor = tensor[0]
+        if tensor.dim() == 4 and tensor.shape[-1] not in (3, 4):
+            if tensor.shape[0] in (3, 4):
+                tensor = tensor.permute(1, 2, 3, 0)
+            elif tensor.shape[1] in (3, 4):
+                tensor = tensor.permute(0, 2, 3, 1)
+        if tensor.is_floating_point():
+            if float(tensor.min()) < 0.0:
+                tensor = tensor.clamp(-1, 1) * 0.5 + 0.5
+            else:
+                tensor = tensor.clamp(0, 1)
+        return tensor.float().numpy()
+    if isinstance(video, np.ndarray):
+        array = video[0] if video.ndim == 5 else video
+        if np.issubdtype(array.dtype, np.integer):
+            return array.astype(np.float32) / 255.0
+        return array.astype(np.float32, copy=False)
+    if isinstance(video, list):
+        return np.stack([_to_frame_array(frame) for frame in video], axis=0)
+    return np.asarray(video, dtype=np.float32)
+
+
+def main() -> None:
+    args = parse_args()
+    generator = torch.Generator(device=current_omni_platform.device_type).manual_seed(args.seed)
+    omni = Omni(
+        model=args.model,
+        model_class_name="LingBotVideoPipeline",
+        flow_shift=args.flow_shift,
+        parallel_config=DiffusionParallelConfig(),
+    )
+
+    prompt: dict[str, Any] = {
+        "prompt": args.prompt,
+        "modalities": ["video"],
+        "num_frames": args.num_frames,
+    }
+    if args.negative_prompt is not None:
+        prompt["negative_prompt"] = args.negative_prompt
+
+    sampling_params = OmniDiffusionSamplingParams(
+        height=args.height,
+        width=args.width,
+        num_frames=args.num_frames,
+        num_inference_steps=args.num_inference_steps,
+        guidance_scale=args.guidance_scale,
+        generator=generator,
+        output_type=args.output_type,
+        extra_args={"batch_cfg": args.batch_cfg, "shift": args.flow_shift},
+    )
+
+    start = time.perf_counter()
+    output = omni.generate(prompt, sampling_params)
+    elapsed = time.perf_counter() - start
+    print(f"Total generation time: {elapsed:.4f} seconds ({elapsed * 1000:.2f} ms)")
+    peak_mb = _peak_memory_mb(output)
+    if peak_mb:
+        print(f"Worker peak GPU memory (reserved): {peak_mb:.2f} MiB ({peak_mb / 1024:.2f} GiB)")
+
+    video = _to_frame_array(_extract_video(output))
+    output_path = Path(args.output)
+    output_path.parent.mkdir(parents=True, exist_ok=True)
+    from diffusers.utils import export_to_video
+
+    export_to_video(list(video), str(output_path), fps=args.fps)
+    print(f"Saved generated video to {output_path}")
+    omni.close()
+
+
+if __name__ == "__main__":
+    main()
@@ -138,7 +138,7 @@ def _extract_peak_memory_mb(result: Any) -> float:
        return 0.0
    val = getattr(result, "peak_memory_mb", 0.0)
    if not val:
-        inner = result
+        inner = getattr(result, "request_output", None)
        if isinstance(inner, list):
            inner = inner[0] if inner else None
        val = getattr(inner, "peak_memory_mb", 0.0)
@@ -1,9 +1,6 @@
 # Image-To-Video

-This example demonstrates online image-to-video generation with vLLM-Omni.
-The startup script defaults to Wan2.2, but `MODEL` can select any supported
-image-to-video model. The existing curl helper remains a concrete Wan2.2
-request through the standard Videos API.
+This example demonstrates how to deploy the Wan2.2 image-to-video model for online video generation using vLLM-Omni.

 ## Start Server

@@ -23,7 +20,9 @@ bash run_server.sh

 The script allows overriding:
 - `MODEL` (default: `Wan-AI/Wan2.2-I2V-A14B-Diffusers`)
- `PORT` (default: `8099`)
+- `PORT` (default: `8091`)
+- `BOUNDARY_RATIO` (default: `0.875`)
+- `FLOW_SHIFT` (default: `12.0`)
 - `CACHE_BACKEND` (default: `none`)
 - `ENABLE_CACHE_DIT_SUMMARY` (default: `0`)

@@ -1,12 +1,12 @@
 #!/bin/bash
-# Shared image-to-video server start script
+# Wan2.2 image-to-video server start script

 MODEL="${MODEL:-Wan-AI/Wan2.2-I2V-A14B-Diffusers}"
 PORT="${PORT:-8099}"
 CACHE_BACKEND="${CACHE_BACKEND:-none}"
 ENABLE_CACHE_DIT_SUMMARY="${ENABLE_CACHE_DIT_SUMMARY:-0}"

-echo "Starting image-to-video server..."
+echo "Starting Wan2.2 I2V server..."
 echo "Model: $MODEL"
 echo "Port: $PORT"
 echo "Cache backend: $CACHE_BACKEND"
@@ -1,9 +1,6 @@
 # Text-To-Image

-This example demonstrates online text-to-image generation with vLLM-Omni. The
-startup script and Python clients default to Qwen-Image, but `MODEL` can select
-any supported text-to-image model. The existing curl helper remains a concrete
-Qwen-Image request through the standard Images API.
+This example demonstrates how to deploy Qwen-Image model for online image generation service using vLLM-Omni.

 ## Start Server

@@ -245,8 +242,8 @@ count, use `size` and `n` rather than `height`, `width`, or
 | `height`                 | int   | None    | Image height in pixels         |
 | `width`                  | int   | None    | Image width in pixels          |
 | `size`                   | str   | None    | Image size (e.g., "1024x1024") |
-| `num_inference_steps`    | int   | model   | Number of denoising steps      |
-| `true_cfg_scale`         | float | model   | True CFG scale when supported  |
+| `num_inference_steps`    | int   | 50      | Number of denoising steps      |
+| `true_cfg_scale`         | float | 4.0     | Qwen-Image CFG scale           |
 | `seed`                   | int   | None    | Random seed (reproducible)     |
 | `negative_prompt`        | str   | None    | Negative prompt                |
 | `num_outputs_per_prompt` | int   | 1       | Number of images to generate   |
@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 """
-Shared text-to-image Gradio demo for online serving.
+Qwen-Image Gradio Demo for online serving.

 Usage:
    python gradio_demo.py [--server http://localhost:8091] [--port 7860]
@@ -77,9 +77,9 @@ def generate_image(
 def create_demo(server_url: str):
    """Create Gradio demo interface."""

-    with gr.Blocks(title="Text-to-Image Demo") as demo:
-        gr.Markdown("# Text-to-Image Online Generation")
-        gr.Markdown("Generate images using the model configured on the server")
+    with gr.Blocks(title="Qwen-Image Demo") as demo:
+        gr.Markdown("# Qwen-Image Online Generation")
+        gr.Markdown("Generate images using Qwen-Image model")

        with gr.Row():
            with gr.Column(scale=1):
@@ -115,6 +115,7 @@ def create_demo(server_url: str):
                        label="Inference Steps",
                        minimum=10,
                        maximum=100,
+                        # Default steps aligned with run_curl_text_to_image.sh to 100
                        value=100,
                        step=5,
                    )
@@ -172,7 +173,7 @@ def create_demo(server_url: str):


 def main():
-    parser = argparse.ArgumentParser(description="Shared text-to-image Gradio demo")
+    parser = argparse.ArgumentParser(description="Qwen-Image Gradio Demo")
    parser.add_argument("--server", default="http://localhost:8091", help="Server URL")
    parser.add_argument("--port", type=int, default=7860, help="Gradio port")
    parser.add_argument("--share", action="store_true", help="Create public link")
@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 """
-Shared OpenAI-compatible text-to-image generation client.
+Qwen-Image OpenAI-compatible image generation client.

 Usage:
    python openai_chat_client.py --prompt "A beautiful landscape" --output output.png
@@ -39,7 +39,7 @@ def generate_image(
        height: Image height in pixels
        width: Image width in pixels
        steps: Number of diffusion steps
-        true_cfg_scale: True CFG scale for models that support it
+        true_cfg_scale: Qwen-Image CFG scale
        seed: Random seed
        negative_prompt: Negative prompt
        num_outputs_per_prompt: Number of images to generate
@@ -70,7 +70,7 @@ def generate_image(
        payload["num_inference_steps"] = steps
    if true_cfg_scale is not None:
        payload["true_cfg_scale"] = true_cfg_scale
-    if negative_prompt is not None:
+    if negative_prompt:
        payload["negative_prompt"] = negative_prompt
    if seed is not None:
        payload["seed"] = seed
@@ -115,15 +115,15 @@ def generate_image(


 def main():
-    parser = argparse.ArgumentParser(description="Shared text-to-image client")
+    parser = argparse.ArgumentParser(description="Qwen-Image chat client")
    parser.add_argument("--prompt", "-p", default="a cup of coffee on the table", help="Text prompt")
-    parser.add_argument("--output", "-o", default="text_to_image_output.png", help="Output file")
+    parser.add_argument("--output", "-o", default="qwen_image_output.png", help="Output file")
    parser.add_argument("--server", "-s", default="http://localhost:8091", help="Server URL")
-    parser.add_argument("--height", type=int, help="Image height; defaults to the model setting")
-    parser.add_argument("--width", type=int, help="Image width; defaults to the model setting")
-    parser.add_argument("--steps", type=int, help="Inference steps; defaults to the model setting")
-    parser.add_argument("--cfg-scale", type=float, help="True CFG scale when supported")
-    parser.add_argument("--seed", type=int, help="Random seed")
+    parser.add_argument("--height", type=int, default=1024, help="Image height")
+    parser.add_argument("--width", type=int, default=1024, help="Image width")
+    parser.add_argument("--steps", type=int, default=50, help="Inference steps")
+    parser.add_argument("--cfg-scale", type=float, default=4.0, help="True CFG scale")
+    parser.add_argument("--seed", type=int, default=0, help="Random seed")
    parser.add_argument("--negative", help="Negative prompt")

    parser.add_argument("--lora-path", default=None, help="Server-local LoRA adapter folder (PEFT format)")
@@ -1,10 +1,10 @@
 #!/bin/bash
-# Shared text-to-image online serving startup script
+# Qwen-Image online serving startup script

 MODEL="${MODEL:-Qwen/Qwen-Image}"
 PORT="${PORT:-8091}"

-echo "Starting text-to-image server..."
+echo "Starting Qwen-Image server..."
 echo "Model: $MODEL"
 echo "Port: $PORT"

@@ -16,8 +16,7 @@ For the full list of supported architectures across all modalities, see
 |---|---|---|---|---|---|
 | Fish Speech S2 Pro | `fishaudio/s2-pro` | ✓ (`ref_audio`+`ref_text`) | ✓ (PCM stream) | — | ✓ |
 | GLM-TTS | `zai-org/GLM-TTS` | ✓ (`ref_audio`+`ref_text`, required) | ✓ (PCM stream) | — | ✓ |
-| IndexTTS-2 | `IndexTeam/IndexTTS-2` | ✓ (`ref_audio` or uploaded `voice`) | `stream=true` response, non-chunk | uploaded audio voice only; no presets | — |
-| IndexTTS-2.5 | native `checkpoints/` bundle | ✓ (`ref_audio` or uploaded `voice`) | `stream=true` response, non-chunk | uploaded audio voice only; no presets | — |
+| IndexTTS-2 | `IndexTeam/IndexTTS-2` | ✓ (`ref_audio` or uploaded `voice`) | compat only, non-chunk | uploaded audio voice only; no presets | — |
 | Ming-omni-tts | `inclusionAI/Ming-omni-tts-0.5B` | ✓ (`ref_audio` / `speaker_embedding`) | ✓ (PCM stream) | IP labels + structured `instructions` | — |
 | Ming-flash-omni-TTS | `Jonathan1909/Ming-flash-omni-2.0` | — (caption-controlled) | — | caption fields (`instructions`) | — |
 | MOSS-TTS-Nano | `OpenMOSS-Team/MOSS-TTS-Nano` | ✓ (`ref_audio` required) | ✓ (PCM stream) | — | ✓ |
@@ -139,29 +138,15 @@ bash examples/online_serving/text_to_speech/glm_tts/run_gradio_demo.sh

 ---

-## IndexTTS-2 and IndexTTS-2.5
+## IndexTTS-2

-2-stage TTS at 22.05 kHz. Requests use `ref_audio` for voice cloning, or an uploaded audio `voice` from `/v1/audio/voices`. IndexTTS-2.5 uses a multilingual tokenizer, CAMPPlus speaker projection, EnhancedCodec, and code-only Stage 0→1 transfer by default. Both versions support emotion conditioning via `emo_audio`, `emo_text`, or `emo_vector` passed in `extra_params`.
-
-### Prerequisites
-
-IndexTTS-2 uses its existing frontend and does not require the IndexTTS-2.5 text dependencies. Before launching IndexTTS-2.5, install its optional frontend dependencies:
-
-```bash
-pip install 'vllm-omni[indextts2]'
-```
+2-stage TTS (GPT AR + S2Mel CFM DiT + BigVGAN) at 22.05 kHz. Requests use `ref_audio` for voice cloning, or an uploaded audio `voice` from `/v1/audio/voices`. Supports emotion conditioning via `emo_audio`, `emo_text`, or `emo_vector` passed in `extra_params`.

 ### Launch
 ```bash
 vllm serve IndexTeam/IndexTTS-2 --omni --trust-remote-code --port 8092
 # or, to pass the bundled deploy config explicitly:
 bash examples/online_serving/text_to_speech/indextts2/run_server.sh
-
-# IndexTTS-2.5 default: use_gpt_latent=false
-MODEL_VERSION=2.5 \
-MODEL=/path/to/indextts-2.5 \
-bash examples/online_serving/text_to_speech/indextts2/run_server.sh
-
 ```

 ### Sending requests
@@ -176,35 +161,14 @@ python examples/online_serving/text_to_speech/indextts2/speech_client.py \
    --text "今天心情很好！" \
    --ref-audio /path/to/ref.wav \
    --emo-audio /path/to/happy.wav
-
-# IndexTTS-2.5 Japanese request
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-    --model-version 2.5 \
-    --model /path/to/indextts-2.5 \
-    --lang ja \
-    --text "こんにちは、音声合成のテストです。" \
-    --ref-audio /path/to/ref.wav
-
-# IndexTTS-2.5 native speed control
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-    --model-version 2.5 \
-    --model /path/to/indextts-2.5 \
-    --speed 2.0 \
-    --text "这是两倍速度的语音合成测试。" \
-    --ref-audio /path/to/ref.wav
 ```

 ### Notes
 - Output: 22.05 kHz mono WAV.
- Provide `ref_audio` on the documented raw request path, or pass `voice` only when it names an uploaded audio voice; neither version provides a built-in text-only preset voice.
- IndexTTS-2.5 request controls `lang` and `text_normalization` are carried in `extra_params`; the client does this for `--model-version 2.5`.
- IndexTTS-2.5 accepts the public `speed` request field in `[0.5, 2.0]`: `2.0` generates shorter, faster speech and `0.5` generates longer, slower speech. This is native Stage 1 duration control, so serving does not apply the generic playback-speed adjustment a second time. IndexTTS-2 does not use this control.
- IndexTTS-2.5 accepts language codes such as `zh`, `en`, `zhen` (mixed Chinese/English), `ja`, and `yue`. `Mandarin` is a vLLM-Omni convenience alias for `zh`. Japanese (`ja`) uses `fugashi` tokenization and produces audio, but it does not automatically expand numbers, dates, or percentages; callers should first write those inputs as readable Japanese text.
- A request `seed` controls Stage 0 AR sampling and per-request CFM noise. Different concurrent batch compositions do not guarantee a bit-identical waveform.
- IndexTTS-2.5 Stage 0 uses plain vLLM sampling and does not reproduce the official default `num_beams=3` beam search. For parity comparisons, run upstream with `num_beams=1`; output quality can differ from the official beam-search result.
+- Provide `ref_audio` on the documented raw request path, or pass `voice` only when it names an uploaded audio voice; IndexTTS-2 does not provide a built-in text-only preset voice.
 - Emotion params (`emo_audio`, `emo_text`, `emo_vector`, `emo_alpha`, `use_emo_text`, `use_random`) are passed via the `extra_params` field. Official precedence is `use_emo_text` > `emo_vector` > `emo_audio` > same emotion as the speaker reference.
- IndexTTS-2.5 uses `vllm_omni/deploy/indextts2_5.yaml` with the official code-only Stage 0 to Stage 1 contract.
- For IndexTTS-2.5, set `MODEL` to the local native bundle; asset discovery accepts its nested `checkpoints/` layout.
+- `stream=true` is accepted as an OpenAI-compatible response path, but IndexTTS-2 is not async-chunk streaming; audio is produced after S2Mel receives the full mel-code sequence.
+- Deploy config: `vllm_omni/deploy/indextts2.yaml` (auto-loaded).

 ---

@@ -605,13 +569,6 @@ python qwen3_tts/streaming_speech_client.py \
 The client writes one PCM file per sentence and a matching
 `sentence_XXX_timestamps.json` sidecar.

-Non-streaming requests can also ask for timestamps: pass
-`"word_timestamps": true` to `POST /v1/audio/speech` and read the
-`X-Word-Timestamps` response header (JSON list of `{word, start_ms, end_ms}`,
-ASCII-escaped). Past 4 KB the header is replaced by
-`X-Word-Timestamps-Omitted: oversize; bytes=<n>; limit=4096` and the audio
-still returns — use the WebSocket path for long transcripts.
-
 To *see* the alignment instead of reading a JSON sidecar, run the
 word-timestamp Gradio demo (server must be launched with `--forced-aligner`):
 ```bash
@@ -1,37 +1,20 @@
 #!/bin/bash
-# Launch vLLM-Omni server for IndexTTS 2.0 or 2.5.
+# Launch vLLM-Omni server for IndexTTS2
 #
 # Usage from repository root:
 #   examples/online_serving/text_to_speech/indextts2/run_server.sh
-#   MODEL_VERSION=2.5 MODEL=/path/to/native/bundle \
-#     examples/online_serving/text_to_speech/indextts2/run_server.sh
+#   CUDA_VISIBLE_DEVICES=0 PORT=8092 MODEL=/path/to/IndexTeam/IndexTTS-2 examples/online_serving/text_to_speech/indextts2/run_server.sh

 set -e

 SCRIPT_DIR="$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")" && pwd)"
 ROOT_DIR="$(cd -- "$SCRIPT_DIR/../../../.." && pwd)"

-MODEL_VERSION="${MODEL_VERSION:-2.0}"
+MODEL="${MODEL:-IndexTeam/IndexTTS-2}"
 PORT="${PORT:-8092}"
+DEPLOY_CONFIG="${DEPLOY_CONFIG:-$ROOT_DIR/vllm_omni/deploy/indextts2.yaml}"

-if [[ "$MODEL_VERSION" == "2.5" ]]; then
-    if [[ -z "${MODEL:-}" ]]; then
-        echo "Usage: MODEL_VERSION=2.5 MODEL=/path/to/native/bundle bash $0" >&2
-        exit 2
-    fi
-    DEFAULT_DEPLOY_CONFIG="$ROOT_DIR/vllm_omni/deploy/indextts2_5.yaml"
-elif [[ "$MODEL_VERSION" == "2.0" ]]; then
-    MODEL="${MODEL:-IndexTeam/IndexTTS-2}"
-    DEFAULT_DEPLOY_CONFIG="$ROOT_DIR/vllm_omni/deploy/indextts2.yaml"
-else
-    echo "MODEL_VERSION must be 2.0 or 2.5" >&2
-    exit 2
-fi
-
-DEPLOY_CONFIG="${DEPLOY_CONFIG:-$DEFAULT_DEPLOY_CONFIG}"
-
-echo "Starting IndexTTS $MODEL_VERSION server with model: $MODEL"
-echo "Deploy config: $DEPLOY_CONFIG"
+echo "Starting IndexTTS2 server with model: $MODEL"

 FLASHINFER_DISABLE_VERSION_CHECK=1 \
 vllm serve "$MODEL" \
@@ -1,6 +1,6 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""OpenAI-compatible client for IndexTTS 2.0/2.5 via /v1/audio/speech.
+"""OpenAI-compatible client for IndexTTS2 TTS via /v1/audio/speech endpoint.

 Examples:
    # With reference audio for voice cloning
@@ -54,44 +54,13 @@ def main() -> None:
    parser.add_argument("--emo-alpha", type=float, default=None, help="Emotion weight in [0, 1]")
    parser.add_argument("--use-emo-text", action="store_true", help="Infer emotion vector from emo-text or text")
    parser.add_argument("--use-random", action="store_true", help="Use random emotion prototypes")
-    parser.add_argument(
-        "--model-version",
-        choices=("2.0", "2.5"),
-        default="2.0",
-    )
-    parser.add_argument("--model", type=str, default=None)
-    parser.add_argument(
-        "--lang",
-        default="zh",
-        help=(
-            "IndexTTS 2.5 language code, for example zh/en/zhen/ja/yue; "
-            "zhen is mixed Chinese/English and Mandarin is a vLLM-Omni alias for zh"
-        ),
-    )
-    parser.add_argument(
-        "--no-text-normalization",
-        action="store_false",
-        dest="text_normalization",
-        help="Disable IndexTTS 2.5 text normalization",
-    )
-    parser.add_argument(
-        "--speed",
-        type=float,
-        default=1.0,
-        help="IndexTTS 2.5 native synthesis speed in [0.5, 2.0]; 2.0 is faster",
-    )
+    parser.add_argument("--model", type=str, default="IndexTeam/IndexTTS-2")
    parser.add_argument("--voice", type=str, default=None, help="Uploaded voice name to use instead of --ref-audio")
    parser.add_argument("--output", type=str, default="output.wav")
    parser.add_argument("--api-base", type=str, default=DEFAULT_API_BASE)
    parser.add_argument("--api-key", type=str, default=DEFAULT_API_KEY)
    parser.add_argument("--response-format", type=str, default="wav")
    args = parser.parse_args()
-    if args.model is None:
-        if args.model_version == "2.5":
-            parser.error("--model is required for IndexTTS 2.5")
-        args.model = "IndexTeam/IndexTTS-2"
-    if args.model_version == "2.5" and not 0.5 <= args.speed <= 2.0:
-        parser.error("IndexTTS 2.5 --speed must be between 0.5 and 2.0")

    if not args.ref_audio and not args.voice:
        parser.error("IndexTTS2 requires --ref-audio or --voice for voice cloning")
@@ -128,10 +97,6 @@ def main() -> None:
        extra_params["use_emo_text"] = True
    if args.use_random:
        extra_params["use_random"] = True
-    if args.model_version == "2.5":
-        payload["speed"] = args.speed
-        extra_params["lang"] = args.lang
-        extra_params["text_normalization"] = args.text_normalization
    if extra_params:
        payload["extra_params"] = extra_params

@@ -73,8 +73,6 @@ plugins:
  - awesome-nav
  - redirects:
      redirect_maps:
-        community/meetups.md: community/meetings.md
-        community/volunteers.md: community/contact_us.md
        user_guide/diffusion/request_batching.md: user_guide/diffusion/execution_modes.md
        user_guide/diffusion/step_execution.md: user_guide/diffusion/execution_modes.md
        user_guide/diffusion/cpu_offload_diffusion.md: user_guide/diffusion/cpu_offload.md
@@ -113,7 +111,6 @@ plugins:
        - "vllm_omni.entrypoints.openai"  # avoid importing vllm in mkdocs building
        - "vllm_omni.entrypoints.openai.protocol"  # avoid importing vllm in mkdocs building
        - "vllm_omni.experimental"  # avoid importing optional serving deps (openai, fastapi, uvicorn, httpx) in mkdocs building
-        - "vllm_omni.model_executor.models.nemotron_voicechat.nemo_vendored"  # vendored NeMo code; upstream docstrings trip griffe strict warnings
  - mkdocstrings:
      handlers:
        python:
@@ -84,13 +84,6 @@ forced-aligner = [
    "qwen-asr>=0.0.6",  # faithful Japanese/Korean word segmentation for the TTS forced aligner
 ]

-indextts2 = [
-    "fugashi>=1.2.0",
-    "unidic-lite>=1.0.0",
-    "wetext>=0.0.9; sys_platform != 'linux'",
-    "WeTextProcessing; sys_platform == 'linux'",
-]
-
 # Local-test-only extras
 local = [
    "FlagEmbedding",
@@ -206,12 +199,6 @@ ignore = [
 # These files keep upstream conventions (star imports, long lines, math-symbol
 # argument names) so they stay easy to diff against upstream releases.
 "vllm_omni/model_executor/models/minicpmo_4_5/minicpmo_4_5_omni_llm.py" = ["E402", "E501", "F403", "F405", "N803"]
-# NemotronVoiceChat inference modules vendored from NVIDIA-NeMo/Speech (branch
-# nemotron-labs-voicechat). Numerics-bearing code is kept byte-close to
-# upstream to stay diffable and parity-safe, so upstream style (long lines,
-# legacy typing, whitespace, unused-import shims) is exempted. TID251 and the
-# F-correctness rules still apply.
-"vllm_omni/model_executor/models/nemotron_voicechat/nemo_vendored/**" = ["E501", "E721", "E741", "F401", "F841", "I001", "N803", "N814", "UP006", "UP007", "UP008", "UP011", "UP032", "UP035", "W291", "W293"]
 "vllm_omni/experimental/fullduplex/joyvl/decision/prompts.py" = ["E501"]  # System/summarizer prompts are long prose lines

 [tool.ruff.lint.flake8-tidy-imports.banned-api]
@@ -320,10 +307,6 @@ extend-ignore-identifiers-re = [
    ".*[Nn][Dd]3.*",
    ".*updator.*",
    "byt5.*",
-    # NemotronVoiceChat vendored triton kernel: dout/DOUT abbreviate d_out
-    # (output feature dim), not "doubt".
-    ".*dout.*",
-    ".*DOUT.*",
 ]
 extend-ignore-re = [
    "\\.\\.\\.n,d->\\.\\.\\.nd",
@@ -344,10 +327,3 @@ byt5 = "byt5"
 # "seeked" is the standard HTML media element DOM event name (fires when a
 # seek completes), used by the word-timestamps Gradio demo. Not a typo.
 seeked = "seeked"
-# Ryan Leary: author credited in the NemotronVoiceChat vendored NeMo
-# features.py copyright header. A surname, not a typo of "leery".
-Leary = "Leary"
-# IndexTTS 2.5 language codes and speech-emotion-recognition event token.
-fo = "fo"
-ba = "ba"
-SER = "SER"
@@ -1,308 +0,0 @@
-# IndexTTS-2.5 for multilingual voice-cloned TTS on 1x GPU
-
-## Summary
-
- Vendor: IndexTeam
- Model: `IndexTeam/IndexTTS-2.5`
- Task: Multilingual text-to-speech with voice cloning, emotion control, and
-  native speed control
- Mode: Online serving with the OpenAI-compatible `/v1/audio/speech` API, or
-  offline inference through `Omni`
- Maintainer: Community
-
-## When to use this recipe
-
-Use this recipe to serve IndexTTS-2.5 as a two-stage TTS system. Stage 0 is an
-autoregressive talker; Stage 1 uses EnhancedCodec, S2Mel CFM/DiT, and BigVGAN
-to produce 22.05 kHz mono speech. Each request supplies synthesis text and
-reference audio, or an uploaded audio voice, for zero-shot voice cloning.
-
-IndexTTS-2.5 adds multilingual text processing and model-native speed control
-to the IndexTTS-2 serving contract.
-
-## References
-
- Upstream model:
-  [IndexTeam/IndexTTS-2.5 on Hugging Face](https://huggingface.co/IndexTeam/IndexTTS-2.5)
- Online serving example:
-  [`examples/online_serving/text_to_speech/README.md#indextts-2-and-indextts-25`](../../examples/online_serving/text_to_speech/README.md#indextts-2-and-indextts-25)
- Offline inference example:
-  [`examples/offline_inference/text_to_speech/README.md#indextts-2-and-indextts-25`](../../examples/offline_inference/text_to_speech/README.md#indextts-2-and-indextts-25)
- OpenAI-compatible client:
-  [`examples/online_serving/text_to_speech/indextts2/speech_client.py`](../../examples/online_serving/text_to_speech/indextts2/speech_client.py)
- Standard deploy config:
-  [`vllm_omni/deploy/indextts2_5.yaml`](../../vllm_omni/deploy/indextts2_5.yaml)
-
-## Hardware Support
-
-### GPU
-
-### 1x NVIDIA H20 96GB
-
-#### Environment
-
- OS: Linux
- Python: 3.10+
- Driver / runtime: NVIDIA CUDA environment
- vLLM version: Match the repository requirements for your checkout
- vLLM-Omni version or commit: Use the commit you are deploying from
-
-Install vLLM-Omni with the IndexTTS text-processing dependencies:
-
-```bash
-pip install 'vllm-omni[indextts2]'
-```
-
-Obtain the native IndexTTS-2.5 bundle and point `MODEL` at its root. The model
-loader accepts the upstream nested `checkpoints/` layout.
-
-#### Command
-
-Start the standard code-only server from the repository root:
-
-```bash
-MODEL_VERSION=2.5 \
-MODEL=/path/to/indextts-2.5 \
-bash examples/online_serving/text_to_speech/indextts2/run_server.sh
-```
-
-This selects `vllm_omni/deploy/indextts2_5.yaml`, which uses
-`use_gpt_latent=false`. To launch directly:
-
-```bash
-vllm serve /path/to/indextts-2.5 \
-  --omni \
-  --trust-remote-code \
-  --port 8092 \
-  --deploy-config vllm_omni/deploy/indextts2_5.yaml
-```
-
-#### Optional: NVIDIA MPS for higher single-GPU throughput
-
-Stage 0 and Stage 1 run in separate processes on the same GPU. Enabling
-NVIDIA Multi-Process Service (MPS) can improve steady-state throughput by
-allowing work from both stages to overlap more effectively. MPS is an opt-in
-deployment optimization: this recipe does not start it automatically, and the
-gain depends on the request mix and concurrency, so benchmark it with the
-intended workload before enabling it in production.
-
-Pay attention to CUDA device renumbering. For example, if the MPS control
-daemon is started with physical `CUDA_VISIBLE_DEVICES=1`, that GPU is exposed
-to MPS clients as logical device `0`; launch the server with
-`CUDA_VISIBLE_DEVICES=0`, not `1`. Use a dedicated MPS pipe/log directory when
-the host is shared with other services, and stop only the MPS daemon owned by
-this deployment.
-
-#### Verification
-
-Send a multilingual voice-cloning request with the bundled client. Local audio
-paths are converted to base64 data URLs before transmission:
-
-```bash
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-  --api-base http://localhost:8092 \
-  --model-version 2.5 \
-  --model /path/to/indextts-2.5 \
-  --lang zh \
-  --text "你好，这是 IndexTTS-2.5 语音合成测试。" \
-  --ref-audio /path/to/reference.wav \
-  --output indextts2_5.wav
-```
-
-##### Native speed control
-
-The public `speed` field is handled natively. Its accepted range is
-`[0.5, 2.0]`; values above `1.0` produce shorter, faster speech. The serving
-adapter converts the API convention to the model-native control with
-`duration_factor = 1.0 / speed`, so `speed=1.25` uses
-`duration_factor=0.8`:
-
-```bash
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-  --api-base http://localhost:8092 \
-  --model-version 2.5 \
-  --model /path/to/indextts-2.5 \
-  --speed 1.25 \
-  --text "这是使用模型原生语速控制的测试。" \
-  --ref-audio /path/to/reference.wav \
-  --output indextts2_5_speed.wav
-```
-
-##### Upload and reuse a named voice
-
-Upload a reference recording once when callers should reuse the same voice
-without sending `ref_audio` on every request. `consent` is required by the
-voice-storage API:
-
-```bash
-curl -X POST http://localhost:8092/v1/audio/voices \
-  -F "audio_sample=@/path/to/reference.wav" \
-  -F "consent=user-consent-id" \
-  -F "name=indextts_demo_voice" \
-  -F "speaker_description=IndexTTS-2.5 demonstration voice"
-```
-
-Confirm that the voice was stored, then synthesize with its name instead of a
-reference-audio payload:
-
-```bash
-curl http://localhost:8092/v1/audio/voices
-
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-  --api-base http://localhost:8092 \
-  --model-version 2.5 \
-  --model /path/to/indextts-2.5 \
-  --voice indextts_demo_voice \
-  --lang zh \
-  --text "你好，这是复用已上传音色的语音合成测试。" \
-  --output indextts2_5_named_voice.wav
-```
-
-There are no built-in text-only preset voices. Each named IndexTTS voice must
-first be created from an uploaded reference recording.
-
-##### Multilingual synthesis and text normalization
-
-The client accepts `zh` (Mandarin), `en` (English), `zhen` (mixed Chinese and
-English), `ja` (Japanese), and `yue` (Cantonese). The following commands use
-the same reference voice across all supported language modes:
-
-```bash
-CLIENT=examples/online_serving/text_to_speech/indextts2/speech_client.py
-MODEL=/path/to/indextts-2.5
-REF_AUDIO=/path/to/reference.wav
-
-python "$CLIENT" --model-version 2.5 --model "$MODEL" \
-  --ref-audio "$REF_AUDIO" --lang zh \
-  --text "你好，这是中文语音合成测试。" --output indextts2_5_zh.wav
-
-python "$CLIENT" --model-version 2.5 --model "$MODEL" \
-  --ref-audio "$REF_AUDIO" --lang en \
-  --text "Hello, this is an English speech synthesis test." \
-  --output indextts2_5_en.wav
-
-python "$CLIENT" --model-version 2.5 --model "$MODEL" \
-  --ref-audio "$REF_AUDIO" --lang zhen \
-  --text "Hello，欢迎使用 IndexTTS 二点五。" --output indextts2_5_zhen.wav
-
-python "$CLIENT" --model-version 2.5 --model "$MODEL" \
-  --ref-audio "$REF_AUDIO" --lang ja \
-  --text "こんにちは、これは日本語の音声合成テストです。" \
-  --output indextts2_5_ja.wav
-
-python "$CLIENT" --model-version 2.5 --model "$MODEL" \
-  --ref-audio "$REF_AUDIO" --lang yue \
-  --text "你好，呢段係粵語語音合成測試。" --output indextts2_5_yue.wav
-```
-
-Text normalization is enabled by default. Disable it only when the input is
-already written exactly as it should be spoken:
-
-```bash
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-  --model-version 2.5 --model /path/to/indextts-2.5 \
-  --ref-audio /path/to/reference.wav --lang en --no-text-normalization \
-  --text "The temperature is twenty five degrees." \
-  --output indextts2_5_pre_normalized.wav
-```
-
-For raw HTTP requests, put the same language and text-normalization controls
-under `extra_params`:
-
-```bash
-curl http://localhost:8092/v1/audio/speech \
-  -H "Content-Type: application/json" \
-  -d '{
-    "model": "/path/to/indextts-2.5",
-    "input": "This is an IndexTTS-2.5 test.",
-    "response_format": "wav",
-    "speed": 1.0,
-    "ref_audio": "data:audio/wav;base64,<BASE64_ENCODED_AUDIO>",
-    "extra_params": {
-      "lang": "en",
-      "text_normalization": true
-    }
-  }' \
-  --output indextts2_5_en.wav
-```
-
-##### Emotion control
-
-IndexTTS-2.5 supports an explicit eight-value emotion vector, an emotion text
-description, or a separate emotion-reference recording. Use one source at a
-time. Vector values are ordered as `happy`, `angry`, `sad`, `afraid`,
-`disgusted`, `melancholic`, `surprised`, and `calm`; each value must be in
-`[0, 1.2]`. `emo_alpha` controls the strength in `[0, 1]`.
-
-Use an explicit emotion vector for deterministic control:
-
-```bash
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-  --model-version 2.5 --model /path/to/indextts-2.5 \
-  --ref-audio /path/to/reference.wav --lang zh \
-  --text "今天真是令人开心的一天！" \
-  --emo-vector 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 \
-  --emo-alpha 0.8 --output indextts2_5_happy_vector.wav
-```
-
-Infer the emotion from a natural-language description with `use_emo_text`:
-
-```bash
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-  --model-version 2.5 --model /path/to/indextts-2.5 \
-  --ref-audio /path/to/reference.wav --lang zh \
-  --text "我们终于完成了这项工作！" \
-  --use-emo-text --emo-text "开心、兴奋而且充满活力" \
-  --emo-alpha 0.8 --output indextts2_5_happy_text.wav
-```
-
-Transfer emotion from a recording while retaining the speaker identity from
-`ref_audio`:
-
-```bash
-python examples/online_serving/text_to_speech/indextts2/speech_client.py \
-  --model-version 2.5 --model /path/to/indextts-2.5 \
-  --ref-audio /path/to/reference.wav --lang zh \
-  --text "请用情绪参考音频中的表达方式朗读这句话。" \
-  --emo-audio /path/to/emotion_reference.wav \
-  --emo-alpha 0.8 --output indextts2_5_emotion_audio.wav
-```
-
-If multiple emotion sources are supplied, the official precedence is
-`use_emo_text` > `emo_vector` > `emo_audio` > the emotion in the speaker
-reference. `--use-random` is also available for exploratory random emotion
-prototypes, but should not be used when reproducible conditioning is required.
-
-For offline inference:
-
-```bash
-python examples/offline_inference/text_to_speech/indextts2/end2end.py \
-  --model /path/to/indextts-2.5 \
-  --model-version 2.5 \
-  --lang zh \
-  --speed 1.0 \
-  --text "你好，这是离线语音合成测试。" \
-  --ref-audio /path/to/reference.wav
-```
-
-#### Notes
-
- Hardware scope: the standard recipe has been exercised on one NVIDIA H200.
-  It does not claim an unverified throughput or quality result.
- Audio output: 22.05 kHz mono WAV.
- Voice cloning: reference audio is required on the documented raw request
-  path. Alternatively, `voice` may name an uploaded audio voice; there is no
-  built-in text-only preset voice.
- Native speed: serving maps `speed` to the model's duration factor and skips
-  generic waveform speed adjustment, so speed is applied exactly once.
- Languages: common codes include `zh`, `en`, `zhen` (mixed Chinese/English),
-  `ja`, and `yue`; `Mandarin` is accepted as an alias for `zh`.
- Japanese normalization: `ja` tokenization does not automatically expand
-  numbers, dates, or percentages. Write these inputs as readable Japanese text
-  before inference.
- Emotion controls: `use_emo_text`, `emo_vector`, and `emo_audio` are
-  alternative conditioning modes. Their precedence is `use_emo_text` >
-  `emo_vector` > `emo_audio` > the speaker-reference emotion.
- Sampling difference: Stage 0 uses plain vLLM sampling, not the upstream
-  default `num_beams=3` beam search. Use upstream `num_beams=1` for parity
-  comparisons; output may differ from the official beam-search result.
@@ -4,20 +4,11 @@

 ## Pipelines

-| `--model-class-name` | Task | Required checkpoint repositories |
-|---|---|---|
-| `LTX2Pipeline` | LTX-2 one-stage T2V/I2V | `Lightricks/LTX-2` |
-| `LTX2TwoStagePipeline` | LTX-2 ordinary two-stage T2V/I2V | `Lightricks/LTX-2` |
-| `LTX2DistilledPipeline` | LTX-2 full-distilled two-stage T2V/I2V | `rootonchair/LTX-2-19b-distilled` |
-| `LTX2Pipeline` | LTX-2.3 one-stage T2V/I2V | `diffusers/LTX-2.3-Diffusers` |
-| `LTX2TwoStagePipeline` | LTX-2.3 ordinary two-stage T2V/I2V | `diffusers/LTX-2.3-Diffusers`<br>`Lightricks/LTX-2.3` |
-| `LTX2DistilledPipeline` | LTX-2.3 full-distilled two-stage T2V/I2V | `diffusers/LTX-2.3-Distilled-Diffusers`<br>`Lightricks/LTX-2.3` |
-
-Repositories in the table are download units. A full pipeline repository
-contains the Transformer, text encoder, connectors, VAEs, vocoder, scheduler,
-and tokenizer; an additional repository supplies LoRA or upsampler sidecars.
-The runtime first searches the model root for sidecars under their official
-filenames, then downloads them from the matching Lightricks Hub repository.
+| Model | Checkpoint | Task | `--model-class-name` | Batching |
+|---|---|---|---|---|
+| LTX-2 | `Lightricks/LTX-2` | One-stage T2V/I2V | `LTX2Pipeline` | Yes |
+| LTX-2 distilled | `rootonchair/LTX-2-19b-distilled` | Two-stage T2V/I2V | `LTX2DistilledPipeline` | No |
+| LTX-2.3 | `diffusers/LTX-2.3-Diffusers` | One-stage T2V/I2V | `LTX2Pipeline` | Yes |

 `LTX2Pipeline` is the unified one-stage entry. Checkpoint metadata selects the
 LTX-2 or LTX-2.3 profile; omitting an image selects T2V, while one initial
@@ -25,10 +16,10 @@ image selects I2V. Both one-stage repositories declare this class, so
 `--model-class-name` is optional. LTX-2.3 requires the Diffusers checkpoint;
 the raw `Lightricks/LTX-2.3` safetensors repository is not directly loadable.

-`LTX2TwoStagePipeline` samples the regular model at half resolution, upsamples,
-then refines with the distilled LoRA. `LTX2DistilledPipeline` uses a fully
-merged distilled Transformer in both stages. Both entries support T2V and I2V;
-select their class explicitly.
+`LTX2DistilledPipeline` is the unified distilled entry: omit `image` for T2V
+or provide one initial image for I2V. As in the official LTX pipelines,
+distilled mode is selected explicitly rather than inferred from the checkpoint
+name or metadata; always pass `--model-class-name LTX2DistilledPipeline`.

 ## API Migration

@@ -77,28 +68,24 @@ The model recipe also supplies the default negative prompt. Top-level
 API defaults `num_frames` to `1`, so set it explicitly; the offline LTX scripts
 default to `121`.

-## Two-Stage Defaults
+## Distilled Two-Stage Defaults

-| Parameter | Ordinary | Full-distilled |
-|---|---:|---:|
-| Final width × height | 1536 × 1024 | 1536 × 1024 |
-| Stage 1 width × height | 768 × 512 | 768 × 512 |
-| Frames / frame rate | 121 / 24 | 121 / 24 |
-| Stage 1 / Stage 2 steps | 40 (LTX-2) or 30 (LTX-2.3) / 3 | 8 / 3 |
-| Guidance | Stage 1 guided; Stage 2 positive-only | Fixed positive-only |
+| Parameter | Value |
+|---|---:|
+| Final width × height | 1536 × 1024 |
+| Stage 1 width × height | 768 × 512 |
+| Frames / frame rate | 121 / 24 |
+| Stage 1 / Stage 2 denoise steps | 8 / 3 |
+| Guidance | Fixed positive-only |

-API dimensions are final dimensions and must be divisible by 64. All LTX
-requests require `num_frames = 8k+1`. Ordinary Stage 1 uses the LTX-2 or
-LTX-2.3 one-stage defaults shown above. Distilled schedules are fixed, so
-`num_inference_steps`, when supplied, must be `8`. Both entries reject custom
-sigmas and input latents.
-
-Ordinary two-stage uses layer-fused LoRA for an unquantized BF16 Transformer
-and automatically switches to dynamic LoRA when quantization is enabled.
+The dimensions passed to the API are the final output dimensions and must be
+divisible by 64. Both one- and two-stage requests require `num_frames = 8k+1`.
+The distilled recipe fixes both sigma schedules; `num_inference_steps` must be
+`8` when supplied.

 ## Serving

-Start either one-stage checkpoint:
+Start either checkpoint:

 ```bash
 vllm serve Lightricks/LTX-2 --omni --stage-init-timeout 600
@@ -108,19 +95,11 @@ vllm serve Lightricks/LTX-2 --omni --stage-init-timeout 600
 vllm serve diffusers/LTX-2.3-Diffusers --omni --stage-init-timeout 600
 ```

-Start a two-stage checkpoint with its explicit class:
+Start the distilled checkpoint with its two-stage entry:

 ```bash
 vllm serve rootonchair/LTX-2-19b-distilled --omni \
  --model-class-name LTX2DistilledPipeline --stage-init-timeout 600
-# LTX-2.3 full-distilled; the v1.1 x2 upsampler is resolved when absent
-vllm serve diffusers/LTX-2.3-Distilled-Diffusers --omni \
-  --model-class-name LTX2DistilledPipeline --stage-init-timeout 600
-# Ordinary LTX-2.3 two-stage
-vllm serve diffusers/LTX-2.3-Diffusers --omni \
-  --model-class-name LTX2TwoStagePipeline \
-  --enable-layerwise-offload \
-  --stage-init-timeout 600
 ```

 The same server handles T2V and I2V. A T2V request using the selected recipe's
@@ -156,9 +135,8 @@ together with `input_reference`.

 ## Guidance

-One-stage and ordinary Stage 1 support independent video/audio CFG,
-spatio-temporal guidance (STG), cross-modality guidance, and rescaling.
-Distilled stages and ordinary Stage 2 are fixed positive-only.
+LTX supports independent video/audio CFG, spatio-temporal guidance (STG),
+cross-modality guidance, and rescaling:

 | Parameter | Default | Effect | Alias |
 |---|---:|---|---|
@@ -194,9 +172,9 @@ complete guidance plan in parallel. Despite the legacy `cfg` name, it covers
 text CFG, STG, and cross-modality guidance passes; guidance rescaling is
 applied after all pass predictions are gathered.

-The default one-stage recipe and ordinary Stage 1 have four Transformer passes
-per denoise step: `cond`, `uncond`, `ptb` (STG), and `mod`
-(cross-modality). The useful balanced configurations are therefore:
+The default one-stage recipe has four Transformer passes per denoise step:
+`cond`, `uncond`, `ptb` (STG), and `mod` (cross-modality). The useful balanced
+configurations are therefore:

 | `--cfg-parallel-size` | Passes per rank | Guidance-slot utilization | Notes |
 |---:|---:|---:|---|
@@ -217,8 +195,9 @@ vllm serve diffusers/LTX-2.3-Diffusers --omni \
 ```

 The total device count is the product of `cfg_parallel_size` and the other
-configured parallel dimensions. Positive-only distilled phases do not benefit
-from `cfg_parallel_size > 1`.
+configured parallel dimensions. Distilled pipelines use positive-only
+guidance, so `cfg_parallel_size > 1` does not provide guidance speedup for
+them.

 ### Python API

@@ -263,22 +242,22 @@ noted below.
 | `image` | image or batch, `None` | Direct value wins over request images; no image selects T2V. I2V accepts one image per prompt, and a batch cannot mix T2V/I2V. |
 | `prompt` | string or list, `None` | Positive-text fallback; request prompts win. Mutually exclusive with `prompt_embeds`. |
 | `negative_prompt` | string or list, `None` | Fallback after request values and before the recipe default; mutually exclusive with negative embeddings. |
-| `height` | `int`, `None` | Request → direct value → recipe default; divisible by 32 one-stage or 64 two-stage. |
+| `height` | `int`, `None` | Request → direct value → recipe default; divisible by 32 one-stage or 64 distilled. |
 | `width` | `int`, `None` | Same precedence and alignment as `height`. |
 | `num_frames` | `int`, `None` | Request → direct value → recipe default; must be `8k+1` and also determines audio duration with `frame_rate`. |
 | `frame_rate` | `float`, `None` | Request `frame_rate` → request `fps` → direct value → recipe default. |
-| `num_inference_steps` | `int`, `None` | Request → direct value → recipe default; minimum 2 one-stage and ordinary Stage 1, fixed at 8 for distilled Stage 1. Custom one-stage `sigmas` determine actual steps. |
+| `num_inference_steps` | `int`, `None` | Request → direct value → recipe default; minimum 2 one-stage, fixed at 8 for distilled Stage 1. Custom one-stage `sigmas` determine actual steps. |
 | `sigmas` | list of float, `None` | One-stage only. Request values win; every request in a fused batch must use the same schedule. |
 | `timesteps` | list of int, `None` | Compatibility slot; LTX accepts only `None`. Use `sigmas`. |
-| `guidance_scale` | `float`, `None` | One-stage and ordinary Stage 1 common video/audio CFG fallback; an explicit request value wins. Distilled phases use fixed positive-only guidance. |
+| `guidance_scale` | `float`, `None` | One-stage common video/audio CFG fallback; an explicit request value wins. Distilled two-stage uses fixed positive-only guidance. |
 | `guidance_rescale` | `float`, `None` | Accepts only `None` or `0.0`; use the modality rescale fields. |
 | `noise_scale` | `float`, `0.0` | Compatibility slot; LTX accepts only `0.0`. |
 | `num_videos_per_prompt` | `int`, `1` | Output-count fallback; positive request `num_outputs_per_prompt` wins. |
 | `generator` | generator or list, `None` | Explicit RNG; otherwise request generators/seeds are collated. Lists must match the effective output batch. |
-| `latents` | tensor, `None` | One-stage only. Request tensors win; packed `[B, S, C]` and validated unpacked video layouts are accepted. |
-| `audio_latents` | tensor, `None` | One-stage only. Request tensors win and are collated. |
+| `latents` | tensor, `None` | Request tensors win. One-stage accepts packed `[B, S, C]` and validated unpacked video layouts. |
+| `audio_latents` | tensor, `None` | One-stage initial audio latents; request tensors win and are collated. |
 | `prompt_embeds` | tensor, `None` | Precomputed positive conditioning; requires `prompt_attention_mask` and cannot accompany `prompt`. |
-| `negative_prompt_embeds` | tensor, `None` | One-stage and ordinary Stage 1 conditioning; requires its mask and cannot accompany `negative_prompt`. |
+| `negative_prompt_embeds` | tensor, `None` | One-stage precomputed negative conditioning; requires its mask and cannot accompany `negative_prompt`. |
 | `prompt_attention_mask` | tensor, `None` | Mask for positive embeddings; request `prompt_attention_mask` or `attention_mask` wins. |
 | `negative_prompt_attention_mask` | tensor, `None` | Mask for negative embeddings; request negative mask fields win. |
 | `decode_timestep` | float or list, `0.0` | Video-VAE decode timestep; request value wins. Lists may match 1, prompt batch, or output batch. |
@@ -294,13 +273,13 @@ request prompt payload; LTX guidance fields live in sampling `extra_args`.

 ### Recipe-Specific Request Capabilities

-| Override | One-stage | Ordinary two-stage | Distilled two-stage |
-|---|---|---|---|
-| Guidance | Supported | Stage 1 only; Stage 2 is positive-only | Fixed positive-only |
-| Negative prompt/embeddings | Supported | Supported by Stage 1 | Rejected |
-| `num_inference_steps` | Supported | Controls Stage 1; Stage 2 uses 3 | Fixed at 8 for Stage 1; Stage 2 uses 3 |
-| Custom `sigmas` | Supported | Rejected | Rejected; both phases use fixed schedules |
-| Video/audio latents | Supported | Rejected | Rejected |
+| Override | One-stage | Distilled two-stage |
+|---|---|---|
+| Guidance | Supported | Fixed positive-only |
+| Negative prompt/embeddings | Supported | Rejected |
+| `num_inference_steps` | Supported | Fixed at 8 for Stage 1; Stage 2 uses 3 |
+| Custom `sigmas` | Supported | Rejected; both phases use fixed schedules |
+| Video/audio latents | Supported | Rejected |

 These capability checks apply equally to direct `forward` keywords and
 values in `OmniDiffusionSamplingParams`; unsupported values fail instead of
@@ -330,10 +309,6 @@ bundled offline CLI do not currently expose `sigmas`.
  `--max-num-seqs 1`.
 - `--cfg-parallel-size` shards the complete LTX guidance plan, including STG,
  modality guidance, and rescale-compatible prediction gathering.
- Sequence parallelism may pad audio latents. Pure Ulysses masks the padding;
-  Ring cannot, so audio length must be SP-divisible. Use `ring_degree=1` or a
-  divisible request shape.
- Cache-DiT is one-stage only; multi-stage configurations are rejected.

 ## Operational Notes

@@ -348,15 +323,14 @@ bundled offline CLI do not currently expose `sigmas`.
  request parameter.
 - For benchmarks, use `tests/dfx/perf/tests/test_ltx2_vllm_omni.json` with
  `tests/dfx/perf/scripts/run_diffusion_benchmark.py`.
- Ordinary and full-distilled two-stage T2V/I2V are supported; HQ execution
-  remains out of scope.
+- Current two-stage support is limited to the distilled checkpoint; official
+  LTX two-stage/HQ execution remains out of scope.

 ## References

 - <https://huggingface.co/Lightricks/LTX-2>
 - <https://huggingface.co/Lightricks/LTX-2.3>
 - <https://huggingface.co/diffusers/LTX-2.3-Diffusers>
- <https://huggingface.co/diffusers/LTX-2.3-Distilled-Diffusers>
 - [Online video generation](../../docs/user_guide/examples/online_serving/text_to_video.md)
 - [Diffusion execution modes](../../docs/user_guide/diffusion/execution_modes.md)
 - [T2V offline example](../../examples/offline_inference/text_to_video/text_to_video.md)
@@ -1,198 +0,0 @@
-# MiniMax-H3 on RTX PRO 5000 Blackwell GPUs
-
-This recipe runs MiniMax-H3 in BF16 on 72 GiB RTX PRO 5000 Blackwell GPUs. It
-contains the validated two-GPU DLO configuration and the recommended resident
-configurations: TP1 x Ulysses2 with 20 resident layers on two GPUs, TP2 x
-Ulysses2 on four GPUs, and TP4 x Ulysses2 on eight GPUs. The four- and
-eight-GPU routes do not require offload.
-
-## Capacity requirements
-
-| Resource | Requirement |
-| --- | ---: |
-| GPUs | 2, 4, or 8 x RTX PRO 5000 Blackwell |
-| GPU HBM | 72 GiB per GPU |
-| Checkpoint storage | 135 GiB per partition |
-| Available system RAM | 200 GiB minimum |
-| Recommended system RAM | 384 GiB |
-
-`FL2VA` and `Ref2VA` are separate checkpoint partitions. Start one server at a
-time on a host sized for the minimum system-memory requirement.
-
-## PCIe topology and GPU order
-
-RTX PRO 5000 does not provide NVLink. Before starting the server, identify the
-closest PCIe pairs and their NUMA affinity:
-
-```bash
-nvidia-smi topo -m
-nvidia-smi nvlink -s
-```
-
-For two GPUs, select the closest PCIe pair on one NUMA node. For four GPUs,
-select all cards from one NUMA node. On the validated host,
-physical GPUs `(0,1)` and `(2,3)` are the two `PXB` pairs on NUMA node 0. The
-order `CUDA_VISIBLE_DEVICES=0,2,1,3` maps the Ulysses groups to those local
-pairs. For eight GPUs, the validated order is `0,4,1,5,2,6,3,7`, with host
-memory interleaved across NUMA nodes 0 and 1. Do not copy these IDs blindly:
-reproduce the same PCIe and NUMA relationships on the target host.
-
-## Recommended serving configurations
-
-### Two GPUs
-
-Two 72 GiB cards require distributed layerwise offload. The validated route
-uses TP1 x Ulysses2, keeps 20 leading DiT layers resident, and streams
-rank-local weights without AllGather. Eager execution avoids regional-compile
-instability on this offload path.
-
-```bash
-export MODEL_ROOT=/path/to/MiniMax-H3
-export MODEL="${MODEL_ROOT}/FL2VA"
-export PORT=8091
-
-CUDA_VISIBLE_DEVICES=0,1 \
-VLLM_WORKER_MULTIPROC_METHOD=spawn \
-VLLM_OMNI_VIDEO_SYNC_TIMEOUT=1800 \
-numactl --cpunodebind=0 --membind=0 \
-vllm serve "${MODEL}" \
-  --omni \
-  --host 0.0.0.0 \
-  --port "${PORT}" \
-  --trust-remote-code \
-  --num-gpus 2 \
-  --tensor-parallel-size 1 \
-  --usp 2 \
-  --ring 1 \
-  --text-encoder-tp-size 2 \
-  --vae-patch-parallel-size 2 \
-  --vae-parallel-mode tile \
-  --vae-use-tiling \
-  --diffusion-attention-backend CUDNN_ATTN \
-  --enable-distributed-layerwise-offload \
-  --dlo-no-use-allgather \
-  --dlo-resident-layers 20 \
-  --enforce-eager
-```
-
-### Four GPUs
-
-The validated baseline uses TP2 x Ulysses2, text-encoder TP4, VAE patch
-parallelism 4, and explicit cuDNN BF16 attention. Selecting the backend
-explicitly keeps the recipe independent of platform-default backend changes.
-
-```bash
-export MODEL_ROOT=/path/to/MiniMax-H3
-export MODEL="${MODEL_ROOT}/FL2VA"
-export PORT=8091
-
-CUDA_VISIBLE_DEVICES=0,2,1,3 \
-VLLM_WORKER_MULTIPROC_METHOD=spawn \
-VLLM_OMNI_VIDEO_SYNC_TIMEOUT=1800 \
-numactl --cpunodebind=0 --membind=0 \
-vllm serve "${MODEL}" \
-  --omni \
-  --host 0.0.0.0 \
-  --port "${PORT}" \
-  --trust-remote-code \
-  --num-gpus 4 \
-  --tensor-parallel-size 2 \
-  --usp 2 \
-  --ring 1 \
-  --text-encoder-tp-size 4 \
-  --vae-patch-parallel-size 4 \
-  --vae-parallel-mode tile \
-  --vae-use-tiling \
-  --diffusion-attention-backend CUDNN_ATTN
-```
-
-### Eight GPUs
-
-The recommended eight-GPU route uses TP4 x Ulysses2, text-encoder TP8, VAE
-patch parallelism 8, and host-memory interleaving across both NUMA nodes.
-
-```bash
-export MODEL_ROOT=/path/to/MiniMax-H3
-export MODEL="${MODEL_ROOT}/FL2VA"
-export PORT=8091
-
-CUDA_VISIBLE_DEVICES=0,4,1,5,2,6,3,7 \
-VLLM_WORKER_MULTIPROC_METHOD=spawn \
-VLLM_OMNI_VIDEO_SYNC_TIMEOUT=1800 \
-numactl --interleave=0,1 \
-vllm serve "${MODEL}" \
-  --omni \
-  --host 0.0.0.0 \
-  --port "${PORT}" \
-  --trust-remote-code \
-  --num-gpus 8 \
-  --tensor-parallel-size 4 \
-  --usp 2 \
-  --ring 1 \
-  --text-encoder-tp-size 8 \
-  --vae-patch-parallel-size 8 \
-  --vae-parallel-mode tile \
-  --vae-use-tiling \
-  --diffusion-attention-backend CUDNN_ATTN
-```
-
-For the four- and eight-GPU resident routes, do not add `--enforce-eager`.
-Warm the server once before measuring so regional compilation is outside the
-measured request. The two-GPU DLO route intentionally remains eager.
-
-For Ref2VA, stop the FL2VA server and restart the same command with
-`MODEL="${MODEL_ROOT}/Ref2VA"`.
-
-## Target-hardware validation
-
-All three configurations were exercised on a PCIe-only, dual-socket host with
-eight RTX PRO 5000 GPUs. The run used PyTorch 2.11.0+cu130, CUDA 13.0, driver
-580.95.05, 1344x768 output, 124 frames, and two warmups.
-
-The four-GPU route also received a five-step profiling validation for GPU
-balance:
-
-| Measurement | Result |
-| --- | ---: |
-| Maximum externally sampled peak | 69,219 MiB (67.60 GiB) per GPU |
-| T2VA worker-reported peak | 65,314 MiB |
-| First-frame FL2VA worker-reported peak | 66,780 MiB |
-| T2VA maximum GPU kernel-time deviation | 0.56% |
-| First-frame FL2VA maximum GPU kernel-time deviation | 0.09% |
-
-The recommended routes produced the following 50-step results. MiniMax-H3
-requested 50 denoise steps and executed 49 denoise updates, so per-step latency
-is `denoise / 49`.
-
-| GPUs | Workload | Parallelism | E2E (s) | Text encode (s) | Denoise (s) | VAE decode (s) | Per step (ms) | Peak memory (GiB) |
-| ---: | --- | --- | ---: | ---: | ---: | ---: | ---: | ---: |
-| 2 | T2VA | TP1 x Ulysses2, DLO resident 20 | 515.57 | 1.19 | 504.69 | 8.79 | 10,300 | 36.38 |
-| 2 | FL2VA first-frame I2VA | TP1 x Ulysses2, DLO resident 20 | 553.45 | 1.27 | 541.90 | 8.76 | 11,059 | 36.38 |
-| 4 | T2VA | TP2 x Ulysses2 | 284.01 | 0.03 | 278.73 | 4.36 | 5,688 | 67.58 |
-| 4 | FL2VA first-frame I2VA | TP2 x Ulysses2 | 305.72 | 0.28 | 299.85 | 4.36 | 6,119 | 67.58 |
-| 8 | T2VA | TP4 x Ulysses2 | 163.20 | 0.04 | 159.40 | 2.58 | 3,253 | 45.83 |
-| 8 | FL2VA first-frame I2VA | TP4 x Ulysses2 | 171.62 | 0.27 | 167.25 | 2.57 | 3,413 | 45.83 |
-
-Peak memory is the maximum per-GPU value sampled externally with
-`nvidia-smi`. The two-GPU route leaves about 34.7 GiB below the reported
-73,415 MiB device capacity, the four-GPU route leaves about 4.1 GiB, and the
-eight-GPU route leaves about 25.9 GiB. Re-measure memory for longer reference
-inputs, concurrency greater than one, or a different output shape.
-
-## T2VA request example
-
-```bash
-export API_URL="http://127.0.0.1:${PORT}/v1/videos/sync"
-
-curl -sS --max-time 1800 -X POST "${API_URL}" \
-  -F 'prompt=At night, three cats march into a bedroom playing tiny brass instruments, then abruptly file out, with synchronized room ambience.' \
-  -F 'width=1344' \
-  -F 'height=768' \
-  -F 'aspect_ratio=16:9' \
-  -F 'fps=24' \
-  -F 'num_inference_steps=50' \
-  -F 'flow_shift=12' \
-  -F 'seed=1101' \
-  -F 'extra_params={"task":"t2va","duration":5.0,"audio_flow_shift":3.0}' \
-  -o t2va.mp4
-```
@@ -1,113 +0,0 @@
-# NemotronLabs VoiceChat 11B
-
-> Offline speech-to-speech: a 16 kHz user utterance in, the agent's spoken reply
-> (text + 22.05 kHz WAV) out, on a 3-stage vLLM-Omni pipeline.
-
-## Summary
-
- Vendor: NVIDIA
- Model: [`nvidia/NVIDIA-NemotronLabs-VoiceChat-11B`](https://huggingface.co/nvidia/NVIDIA-NemotronLabs-VoiceChat-11B)
- Task: Speech-to-speech voice chat. The model runs a frame-locked 12.5 Hz
-  timeline: a Conformer + NemotronH hybrid-Mamba thinker emits one text token
-  per acoustic frame, a Gemma3-1B EAR-TTS talker turns the text timeline into
-  31-quantizer RVQ code stacks, and an RVQ-VAE codec decodes them to audio.
- Mode: Offline single-turn inference (batch=1). Online/duplex serving,
-  function calling, and batch>1 are documented follow-ups.
- Maintainer: [`@yuekaizhang`](https://github.com/yuekaizhang)
-
-## When to use this recipe
-
-Use this recipe to run a single-turn voice-chat exchange with
-`NVIDIA-NemotronLabs-VoiceChat-11B` on one GPU: you provide a user utterance as
-a WAV file (any sample rate; it is resampled to 16 kHz mono) plus an optional
-spoken-style system prompt, and get back the agent's reply as text and a
-22.05 kHz WAV. The integration is NeMo-free at runtime — the perception
-Conformer, EAR-TTS talker, and RVQ-VAE codec are vendored, dependency-stripped
-NeMo modules (`nemo_vendored/`), so no `nemo_toolkit` install is needed.
-
-## References
-
- Offline example:
-  [`examples/offline_inference/nemotron_voicechat/end2end.py`](../../examples/offline_inference/nemotron_voicechat/end2end.py)
- Model modules (thinker / talker / code2wav / vendored NeMo):
-  [`vllm_omni/model_executor/models/nemotron_voicechat/`](../../vllm_omni/model_executor/models/nemotron_voicechat/)
- Staged pipeline config:
-  [`vllm_omni/deploy/nemotron_labs_voicechat.yaml`](../../vllm_omni/deploy/nemotron_labs_voicechat.yaml)
- Upstream model card:
-  [`nvidia/NVIDIA-NemotronLabs-VoiceChat-11B`](https://huggingface.co/nvidia/NVIDIA-NemotronLabs-VoiceChat-11B)
- Reference implementation: NVIDIA-NeMo/Speech, branch `nemotron-labs-voicechat`
-
-## Pipeline
-
-| stage | arch | dtype | role |
-|---|---|---|---|
-| 0 thinker | `NemotronVoiceChatThinkerForConditionalGeneration` (LLM_AR) | fp32 | WAV + system prompt -> frame-locked text-token timeline (+ function channel) |
-| 1 talker | `NemotronVoiceChatTalker` (LLM_AR) | fp32 | text timeline -> 31-quantizer RVQ code stacks (one per 80 ms frame) |
-| 2 code2wav | `NemotronVoiceChatCode2Wav` (LLM_GENERATION) | fp32 | RVQ-VAE decode -> 22.05 kHz PCM |
-
-Stages 0/1 default to fp32 for exact parity with the NeMo reference
-implementation (greedy decoding matches it token for token on the acceptance
-fixture). The deploy yaml documents a ~2x-faster bf16 thinker option whose
-output stayed within one word of the reference in testing.
-
-## Hardware Support
-
-## GPU
-
-### 1x H100 80GB
-
-#### Environment
-
- OS: Linux
- Python: 3.12
- vLLM version: 0.26.0
- vLLM-Omni version or commit: this PR / current `main`
-
-#### Command
-
-```bash
-# Tokenizer: the checkpoint ships no HF tokenizer; it resolves from the
-# nvidia/NVIDIA-Nemotron-Nano-9B-v2 HF id automatically. For air-gapped runs,
-# point NEMOTRON_VOICECHAT_LLM_PATH at a local snapshot of that repo instead.
-python examples/offline_inference/nemotron_voicechat/end2end.py \
-    --checkpoint /path/to/NVIDIA-NemotronLabs-VoiceChat-11B \
-    --wav /path/to/user_question.wav \
-    --output-dir results/nemotron_voicechat
-```
-
-#### Verification
-
-```bash
-ls results/nemotron_voicechat
-# <stem>_output.txt          the agent reply as text
-# <stem>_output.wav          the agent reply as 22.05 kHz audio
-# <stem>_text_tokens.json    the frame-locked text-token timeline
-```
-
-The reply text should read as a coherent spoken-style answer to the question in
-the input WAV, and the WAV should transcribe to (approximately) the same text
-with any ASR model.
-
-#### Notes
-
- Memory usage: the shipped yaml runs all three stages on one GPU
-  (`gpu_memory_utilization` 0.62 / 0.12 / 0.06); peak usage is dominated by the
-  fp32 thinker. The fp32 default has a hard floor of roughly 43 GB of thinker
-  weights alone (9B backbone + 587M `embed_tokens` + 587M `function_head` +
-  0.6B Conformer), so 48 GB cards cannot run it — use the bf16 thinker option
-  documented in the deploy yaml on anything smaller than an 80 GB part.
- Input sizing: the timeline is frame-locked, so the reply budget IS the input
-  duration. The acoustic channel trails the text channel; if the WAV does not
-  carry enough trailing silence for the reply to finish, the spoken answer is
-  truncated silently. Leave generous trailing silence (a question ending at
-  ~4.5 s truncated in an 8 s WAV but completed cleanly in 16 s); the offline
-  example warns when the text channel is still speaking near the last frame.
- Key flags: sampling is greedy end to end. The thinker is frame-locked —
-  `max_tokens` equals the acoustic frame count with `ignore_eos=True`. Do NOT
-  set `min_tokens` on the thinker: the tokenizer's EOS token is also the
-  frame-locked PAD/silence token, so masking it forces the model to babble
-  instead of pausing.
- The talker's `max_tokens` is 16383 (its stage prompt is one placeholder
-  token, and the stage context is 16384).
- Known limitations: batch=1 only; offline single turn (no online/duplex
-  serving yet); the function-call channel is decoded but not yet acted on.