revert doc deletion

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>

.buildkite/pipeline.yml

@@ -1,87 +0,0 @@
-steps:
-  - label: ":docker: Build image"
-    key: image-build
-    commands:
-      - "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7"
-      - "docker build --file docker/Dockerfile.ci -t vllm-omni-ci ."
-      - "docker tag vllm-omni-ci public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT"
-      - "docker push public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT"
-    agents:
-      queue: "cpu_queue_premerge_us_east_1"
-
-  - label: "Simple Unit Test"
-    commands:
-      - ".buildkite/scripts/simple_test.sh"
-    agents:
-      queue: "cpu_queue_premerge"
-
-  - label: "Diffusion Model Test"
-    timeout_in_minutes: 15
-    depends_on: image-build
-    commands:
-      - pytest -s -v tests/e2e/offline_inference/test_t2i_model.py
-    agents:
-      queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU
-    plugins:
-      - docker#v5.2.0:
-          image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-          always-pull: true
-          propagate-environment: true
-          environment:
-            - "HF_HOME=/fsx/hf_cache"
-          volumes:
-            - "/fsx/hf_cache:/fsx/hf_cache"
-
-  - label: "Omni Model Test"
-    timeout_in_minutes: 15
-    depends_on: image-build
-    commands:
-      - export VLLM_LOGGING_LEVEL=DEBUG
-      - export VLLM_WORKER_MULTIPROC_METHOD=spawn
-      - pytest -s -v tests/e2e/offline_inference/test_qwen2_5_omni.py
-    agents:
-      queue: "gpu_4_queue" # g6.12xlarge instance on AWS, has 4 L4 GPU
-    plugins:
-      - docker#v5.2.0:
-          image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-          always-pull: true
-          propagate-environment: true
-          environment:
-            - "HF_HOME=/fsx/hf_cache"
-          volumes:
-            - "/fsx/hf_cache:/fsx/hf_cache"
-
-  - label: "Omni Model Test with H100"
-    timeout_in_minutes: 20
-    depends_on: image-build
-    commands:
-      - export VLLM_WORKER_MULTIPROC_METHOD=spawn
-      - pytest -s -v tests/e2e/offline_inference/test_qwen3_omni.py tests/e2e/online_serving/test_qwen3_omni.py
-    agents:
-      queue: "mithril-h100-pool"
-    plugins:
-      - kubernetes:
-          podSpec:
-            containers:
-              - image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
-                resources:
-                  limits:
-                    nvidia.com/gpu: 2
-                volumeMounts:
-                  - name: devshm
-                    mountPath: /dev/shm
-                  - name: hf-cache
-                    mountPath: /root/.cache/huggingface
-                env:
-                  - name: HF_HOME
-                    value: /root/.cache/huggingface
-            nodeSelector:
-              node.kubernetes.io/instance-type: gpu-h100-sxm
-            volumes:
-              - name: devshm
-                emptyDir:
-                  medium: Memory
-              - name: hf-cache
-                hostPath:
-                  path: /mnt/hf-cache
-                  type: DirectoryOrCreate

.buildkite/scripts/simple_test.sh

@@ -1,52 +0,0 @@
-#!/usr/bin/env bash
-
-set -euo pipefail
-
-SCRIPT_DIR="$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")" && pwd)"
-REPO_ROOT="$(cd -- "${SCRIPT_DIR}/../.." && pwd)"
-UV_BIN="uv"
-PYTHON_BOOTSTRAP="${PYTHON:-python3}"
-VENV_DIR="${VENV_DIR:-${REPO_ROOT}/.venv-simple-test}"
-TARGET_PY_VERSION="3.12"
-
-if ! command -v "${PYTHON_BOOTSTRAP}" >/dev/null 2>&1; then
-  PYTHON_BOOTSTRAP="python"
-fi
-
-cd "${REPO_ROOT}"
-
-# Ensure pip-installed scripts (including uv) are on PATH
-PYTHON_SCRIPTS="$(${PYTHON_BOOTSTRAP} -c 'import sysconfig; print(sysconfig.get_path("scripts"))')"
-USER_BASE="$(${PYTHON_BOOTSTRAP} -m site --user-base 2>/dev/null || true)"
-PATH="${PYTHON_SCRIPTS}:${PATH}"
-if [[ -n "${USER_BASE}" ]]; then
-  PATH="${USER_BASE}/bin:${PATH}"
-fi
-
-if ! command -v "${UV_BIN}" >/dev/null 2>&1; then
-  "${PYTHON_BOOTSTRAP}" -m pip install --upgrade pip
-  "${PYTHON_BOOTSTRAP}" -m pip install uv
-  hash -r
-fi
-
-TARGET_PY_PATH="$(${UV_BIN} python find "${TARGET_PY_VERSION}" 2>/dev/null | head -n1 || true)"
-if [[ -z "${TARGET_PY_PATH}" ]]; then
-  "${UV_BIN}" python install "${TARGET_PY_VERSION}"
-  TARGET_PY_PATH="$(${UV_BIN} python find "${TARGET_PY_VERSION}" 2>/dev/null | head -n1 || true)"
-fi
-
-if [[ -z "${TARGET_PY_PATH}" ]]; then
-  echo "Failed to locate python ${TARGET_PY_VERSION}" >&2
-  exit 1
-fi
-
-if [[ ! -d "${VENV_DIR}" ]]; then
-  "${UV_BIN}" venv --python "${TARGET_PY_PATH}" "${VENV_DIR}"
-fi
-
-VENV_PYTHON="${VENV_DIR}/bin/python"
-[[ -x "${VENV_PYTHON}" ]] || { echo "Python not found in ${VENV_DIR}"; exit 1; }
-
-"${UV_BIN}" pip install --python "${VENV_PYTHON}" vllm==0.11.0
-"${UV_BIN}" pip install --python "${VENV_PYTHON}" -e ".[dev]"
-"${VENV_PYTHON}" -m pytest -v -s tests/test_omni_llm.py

.github/ISSUE_TEMPLATE/config.yml

@@ -1,5 +1,5 @@
 blank_issues_enabled: false
 contact_links:
   - name: Questions
-    url: https://slack.vllm.ai
-    about: Ask questions and discuss with other vLLM-Omni community members in our #sig-omni slack channel.
+    url: https://github.com/vllm-project/vllm-omni/discussions
+    about: Ask questions and discuss with other vLLM-omni community members

.github/workflows/ci-all-files.yml

@@ -0,0 +1,60 @@
+name: CI All Files
+
+on:
+  workflow_dispatch:
+
+permissions:
+  contents: read
+
+jobs:
+  test-changed:
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.12", "3.13"]
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0 # Need full history to detect changes
+
+      - name: Free disk space
+        run: |
+          # Remove large, preinstalled toolchains we don't need on the runner
+          sudo rm -rf /usr/share/dotnet
+          sudo rm -rf /opt/ghc
+          sudo rm -rf /usr/local/lib/android
+          sudo rm -rf /opt/hostedtoolcache/CodeQL
+          # Prune docker to reclaim space (safe on GH runners)
+          docker system prune -af || true
+          df -h
+
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Prepare environment with uv
+        run: |
+          python -m pip install --upgrade pip uv
+          uv venv --python "${{ matrix.python-version }}" --seed
+          echo "VIRTUAL_ENV=${PWD}/.venv" >> "$GITHUB_ENV"
+          echo "PATH=${PWD}/.venv/bin:$PATH" >> "$GITHUB_ENV"
+          source .venv/bin/activate
+          uv pip install -e ".[dev]"
+
+      - name: Run tests
+        run: |
+          if [ ! -d tests ]; then
+            echo "tests/ directory not found. Skipping pytest run."
+
+
+            exit 0
+          fi
+          TEST_FILES=$(find tests -type f \( -name "test_*.py" -o -name "*_test.py" \))
+          if [ -z "$TEST_FILES" ]; then
+            echo "No pytest-compatible test files detected under tests/. Skipping pytest run."
+            exit 0
+          fi
+          pytest tests/ -v -m "not slow"

.github/workflows/ci.yml

@@ -0,0 +1,124 @@
+name: CI
+
+on:
+  pull_request:
+    branches: [main]
+  workflow_dispatch:
+    inputs:
+      test_changed_only:
+        description: 'Only test changed files (true) or all files (false)'
+        required: false
+        default: 'true'
+        type: choice
+        options:
+          - 'true'
+          - 'false'
+      enable_changed_files_detection:
+        description: 'Enable changed files detection feature'
+        required: false
+        default: 'true'
+        type: choice
+        options:
+          - 'true'
+          - 'false'
+  workflow_call:
+    inputs:
+      test_changed_only:
+        description: 'Only test changed files (true) or all files (false)'
+        required: false
+        type: string
+        default: 'true'
+      enable_changed_files_detection:
+        description: 'Enable changed files detection feature'
+        required: false
+        type: string
+        default: 'true'
+
+permissions:
+  contents: read
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.12", "3.13"]
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Free disk space
+        run: |
+          # Remove large, preinstalled toolchains we don't need on the runner
+          sudo rm -rf /usr/share/dotnet
+          sudo rm -rf /opt/ghc
+          sudo rm -rf /usr/local/lib/android
+          sudo rm -rf /opt/hostedtoolcache/CodeQL
+          # Prune docker to reclaim space (safe on GH runners)
+          docker system prune -af || true
+          df -h
+
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Prepare environment with uv
+        run: |
+          python -m pip install --upgrade pip uv
+          uv venv --python "${{ matrix.python-version }}" --seed
+          echo "VIRTUAL_ENV=${PWD}/.venv" >> "$GITHUB_ENV"
+          echo "PATH=${PWD}/.venv/bin:${PATH}" >> "$GITHUB_ENV"
+          # Activate env to ensure subsequent commands succeed in this step
+          source .venv/bin/activate
+          uv pip install -e ".[dev]"
+
+      - name: Determine test scope
+        id: test-scope
+        env:
+          ENABLE_CHANGED_FILES_DETECTION: ${{ inputs.enable_changed_files_detection || 'true' }}
+          TEST_CHANGED_ONLY: ${{ inputs.test_changed_only || 'true' }}
+        run: |
+          # Determine if we should test only changed files
+          # If detection is disabled, always test all files
+          if [ "$ENABLE_CHANGED_FILES_DETECTION" != "true" ]; then
+            TEST_CHANGED_ONLY="false"
+          fi
+
+          echo "test_changed_only=$TEST_CHANGED_ONLY" >> "$GITHUB_OUTPUT"
+          echo "enable_detection=$ENABLE_CHANGED_FILES_DETECTION" >> "$GITHUB_OUTPUT"
+          echo "Test scope: $TEST_CHANGED_ONLY"
+          echo "Changed files detection: $ENABLE_CHANGED_FILES_DETECTION"
+
+      - name: Get changed test files
+        id: changed-tests
+        if: steps.test-scope.outputs.test_changed_only == 'true' && steps.test-scope.outputs.enable_detection == 'true'
+        env:
+          GITHUB_BASE_REF: ${{ github.event.pull_request.base.sha || '' }}
+          GITHUB_SHA: ${{ github.sha }}
+        run: |
+          chmod +x .github/scripts/detect_changed_tests.sh
+          .github/scripts/detect_changed_tests.sh \
+            "${{ github.event.pull_request.base.sha || github.event.before || '' }}" \
+            "${{ github.sha }}" \
+            "${{ github.event_name }}"
+
+      - name: Run tests (changed files only)
+        if: steps.test-scope.outputs.test_changed_only == 'true' && steps.changed-tests.outputs.has_tests == 'true'
+        run: |
+          pytest "${{ steps.changed-tests.outputs.test_files }}" -v -m "not slow"
+
+      - name: Run tests (all files)
+        if: steps.test-scope.outputs.test_changed_only == 'false' || steps.changed-tests.outputs.has_tests == 'false'
+        run: |
+          if [ ! -d tests ]; then
+            echo "tests/ directory not found. Skipping pytest run."
+            exit 0
+          fi
+          TEST_FILES=$(find tests -type f \( -name "test_*.py" -o -name "*_test.py" \))
+          if [ -z "$TEST_FILES" ]; then
+            echo "No pytest-compatible test files detected under tests/. Skipping pytest run."
+            exit 0
+          fi
+          pytest tests/ -v -m "not slow"

.github/workflows/pre-commit.yml

@@ -3,8 +3,6 @@ name: pre-commit
 on:
   pull_request:
     branches: [main]
-  push:
-    branches: [main]
   workflow_dispatch: # for manual trigger
 
 concurrency:

.gitignore

@@ -156,8 +156,14 @@ cython_debug/
 # cursor
 .cursor/
 
+# docker
+docker/
+
 # scripts
-/scripts/
+scripts/
+
+# tests
+tests/
 
 # PyCharm
 #  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
@@ -169,7 +175,7 @@ cython_debug/
 #  *.iws
 #  .idea/
 
-# vLLM-Omni specific
+# vLLM-omni specific
 # Model files and checkpoints
 checkpoints/
 *.bin
@@ -179,7 +185,6 @@ checkpoints/
 
 # Cache directories
 cache/
-!vllm_omni/diffusion/cache/
 .cache/
 diffusion_cache/
 kv_cache/
@@ -230,5 +235,3 @@ configs/development.yaml
 # Docker
 .dockerignore
 Dockerfile.dev
-discussion
-tmp_test

.pre-commit-config.yaml

@@ -1,6 +1,3 @@
-default_install_hook_types:
-  - pre-commit
-  - commit-msg
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v4.6.0
@@ -33,28 +30,3 @@ repos:
     hooks:
       - id: actionlint
         files: ^\.github/workflows/.*\.ya?ml$
-
-
-  - repo: local
-    hooks:
-    - id: signoff-commit
-      name: Sign-off Commit
-      entry: bash
-      args:
-        - -c
-        - |
-          if ! grep -q "^Signed-off-by: $(git config user.name) <$(git config user.email)>" "$(git rev-parse --git-path COMMIT_EDITMSG)"; then
-            printf "\nSigned-off-by: $(git config user.name) <$(git config user.email)>\n" >> "$(git rev-parse --git-path COMMIT_EDITMSG)"
-          fi
-      language: system
-      verbose: true
-      stages: [commit-msg]
-
-    # Keep `suggestion` last
-    - id: suggestion
-      name: Suggestion
-      entry: bash -c 'echo "To bypass all the pre-commit hooks, add --no-verify to git commit. To skip a specific hook, prefix the commit command with SKIP=<hook-id>."'
-      language: system
-      verbose: true
-      pass_filenames: false
-    # Insert new entries above the `suggestion` entry

CONTRIBUTING.md

@@ -1,3 +1,19 @@
-# Contributing to vLLM-Omni
+# Contributing to vLLM
 
-You may find information about contributing to vLLM-Omni on [Contributing](https://vllm-omni.readthedocs.io/en/latest/contributing/)
+## Getting Started
+
+Setup environment with uv
+```bash
+uv venv --python 3.12 --seed
+source .venv/bin/activate
+```
+
+Pre-commit has been linked to git commit. For run checks locally, please install pre-commit:
+```bash
+uv pip install pre-commit
+```
+
+Then run:
+```bash
+pre-commit run --show-diff-on-failure --color=always --all-files
+```

README.md

@@ -1,74 +1,107 @@
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/logos/vllm-omni-logo.png">
-    <img alt="vllm-omni" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/logos/vllm-omni-logo.png" width=55%>
-  </picture>
-</p>
-<h3 align="center">
-Easy, fast, and cheap omni-modality model serving for everyone
-</h3>
+# vLLM-omni: Multi-modal Extension for vLLM
 
-<p align="center">
-| <a href="https://vllm-omni.readthedocs.io/en/latest/"><b>Documentation</b></a> | <a href="https://discuss.vllm.ai"><b>User Forum</b></a> | <a href="https://slack.vllm.ai"><b>Developer Slack</b></a> |
-</p>
+vLLM-omni is designed to extend vLLM capabilities to support multi-modality model inference and serving, particularly focusing on non-autoregressive structures and non-textual outputs.
 
----
+## 🎯 Overview
 
-*Latest News* 🔥
+Traditional vLLM systems are limited to text-based, autoregressive generation. vLLM-omni addresses this limitation by enabling support for:
 
-- [2025/11] vLLM community officially released [vllm-project/vllm-omni](https://github.com/vllm-project/vllm-omni) in order to support omni-modality models serving.
+- **Multi-modal Models**: Text, image, video, audio, and sensor data processing
+- **Non-autoregressive Architectures**: Diffusion Transformers (DiT) and other parallel generation models
+- **Heterogeneous Outputs**: Beyond traditional text generation to multimodal outputs
 
----
+## 🏗️ Architecture
 
-## About
+vLLM-omni is built on a modular architecture that extends vLLM's core functionality:
 
-[vLLM](https://github.com/vllm-project/vllm) was originally designed to support large language models for text-based autoregressive generation tasks. vLLM-Omni is a framework that extends its support for omni-modality model inference and serving:
 
-- **Omni-modality**: Text, image, video, and audio data processing
-- **Non-autoregressive Architectures**: extend the AR support of vLLM to Diffusion Transformers (DiT) and other parallel generation models
-- **Heterogeneous outputs**: from traditional text generation to multimodal outputs
+## 🚀 Key Features
 
-<p align="center">
-  <picture>
-    <img alt="vllm-omni" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/omni-modality-model-architecture.png" width=55%>
-  </picture>
-</p>
+### Multi-Engine Support
 
-vLLM-Omni is fast with:
+- **Autoregressive Engine**: Traditional text generation with enhanced KV-caching
+- **Diffusion Engine**: Support for DiT models and iterative generation
+- **Hybrid Engine**: Combined AR+DiT processing pipelines
 
-- State-of-the-art AR support by leveraging efficient KV cache management from vLLM
-- Pipelined stage execution overlapping for high throughput performance
-- Fully disaggregation based on OmniConnector and dynamic resource allocation across stages
+### Modality Processing
 
-vLLM-Omni is flexible and easy to use with:
+- **Text**: Advanced tokenization and embedding generation
+- **Image**: Vision encoder integration (CLIP, etc.)
+- **Audio**: Speech processing and audio embedding
 
-- Heterogeneous pipeline abstraction to manage complex model workflows
-- Seamless integration with popular Hugging Face models
-- Tensor, pipeline, data and expert parallelism support for distributed inference
-- Streaming outputs
-- OpenAI-compatible API server
+## 📋 Supported Models
 
-vLLM-Omni seamlessly supports most popular open-source models on HuggingFace, including:
+### AR + Diffusion Transformer (DiT) Models
+- Qwen-omni (Thinker-Talker-Codec structure)
+- HunyunaImage 3.0 (Ongoing)
+- Qwen-Image (Ongoing)
 
-- Omni-modality models (e.g. Qwen-Omni)
-- Multi-modality generation models (e.g. Qwen-Image)
+## 🛠️ Installation
 
-## Getting Started
+### Environment setup
 
-Visit our [documentation](https://vllm-omni.readthedocs.io/en/latest/) to learn more.
+Use Docker to keep consistent basic environment (Optional, Recommended)
+```bash
+docker run --gpus all --ipc=host --network=host -v $source_dir:$container_dir --name $container_name -it nvcr.io/nvidia/pytorch:25.01-py3 bash
+```
 
-- [Installation](https://vllm-omni.readthedocs.io/en/latest/getting_started/installation/)
-- [Quickstart](https://vllm-omni.readthedocs.io/en/latest/getting_started/quickstart/)
-- [List of Supported Models](https://vllm-omni.readthedocs.io/en/latest/models/supported_models/)
+Set up basic uv environment
+```bash
+pip install uv
+uv venv --python 3.12 --seed
+source .venv/bin/activate
+```
 
-## Contributing
+### Installation of vLLM (for users)
 
-We welcome and value any contributions and collaborations.
-Please check out [Contributing to vLLM-Omni](https://vllm-omni.readthedocs.io/en/latest/contributing/) for how to get involved.
+Now we build it based on vLLM v0.11.0. Please install it with command below.
+```bash
+uv pip install vllm==0.11.0 --torch-backend=auto
+```
 
-## Join the Community
-Feel free to ask questions, provide feedbacks and discuss with fellow users of vLLM-Omni in `#sig-omni` slack channel at [slack.vllm.ai](https://slack.vllm.ai) or vLLM user forum at [discuss.vllm.ai](https://discuss.vllm.ai).
+### Installation of vLLM (for developers)
 
-## License
+If you want to check or debug with source code of vLLM, install stable release version of vllm with 0.11.0 from source with pre-built wheel file.
 
-Apache License 2.0, as found in the [LICENSE](./LICENSE) file.
+```bash
+git clone https://github.com/vllm-project/vllm.git
+cd vllm
+git checkout v0.11.0
+```
+Set up environment variables to get pre-built wheels. If there are internet problems, just download the whl file manually. And set VLLM_PRECOMPILED_WHEEL_LOCATION as your local absolute path of whl file.
+```bash
+export VLLM_PRECOMPILED_WHEEL_LOCATION=https://github.com/vllm-project/vllm/releases/download/v0.11.0/vllm-0.11.0-cp38-abi3-manylinux1_x86_64.whl
+```
+Install vllm with command below (If you have no existing PyTorch).
+```bash
+uv pip install --editable .
+```
+
+Install vllm with command below (If you already have PyTorch).
+```bash
+python use_existing_torch.py
+uv pip install -r requirements/build.txt
+uv pip install --no-build-isolation --editable .
+```
+
+### Verification for successful installation of vLLM
+Just run the command below. If no error, it demonstrates that the installation is successful.
+```bash
+python -c "import vllm._C"
+```
+
+### Installation of vLLM-omni
+Install additional requirements for vllm-omni
+```bash
+git clone https://github.com/vllm-project/vllm-omni.git
+cd vllm_omni
+uv pip install -e .
+```
+
+## Run examples (Qwen2.5-omni)
+
+Please check the folder of [examples](examples)
+
+## Further details
+
+For detailed architecture design, see [vllm_omni_design.md](docs/architecture/vllm_omni_design.md) and [high_level_arch_design.md](docs/architecture/high_level_arch_design.md).

docker/Dockerfile.ci

@@ -1,14 +0,0 @@
-ARG VLLM_BASE_IMAGE=vllm/vllm-openai
-ARG VLLM_BASE_TAG=v0.11.0
-FROM ${VLLM_BASE_IMAGE}:${VLLM_BASE_TAG}
-ARG APP_DIR=/workspace/vllm-omni
-WORKDIR ${APP_DIR}
-
-COPY . .
-
-# Install vllm-omni into the same uv-managed Python environment used by the base image.
-RUN uv pip install --python "$(python3 -c 'import sys; print(sys.executable)')" --no-cache-dir ".[dev]"
-
-RUN ln -sf /usr/bin/python3 /usr/bin/python
-
-ENTRYPOINT []

docs/.nav.yml

@@ -1,55 +0,0 @@
-nav:
-- Home: README.md
-- User Guide:
-  - Getting Started:
-    - getting_started/quickstart.md
-    - getting_started/installation
-  - Examples:
-    - examples/README.md
-    - Offline Inference:
-      - Qwen2.5-Omni: user_guide/examples/offline_inference/qwen2_5_omni.md
-      - Qwen3-Omni: user_guide/examples/offline_inference/qwen3_omni.md
-      - Text-To-Image: user_guide/examples/offline_inference/text_to_image.md
-      - Image-To-Image: user_guide/examples/offline_inference/image_to_image.md
-      - Text-To-Video: user_guide/examples/offline_inference/text_to_video.md
-    - Online Serving:
-      - Qwen2.5-Omni: user_guide/examples/online_serving/qwen2_5_omni.md
-      - Qwen3-Omni: user_guide/examples/online_serving/qwen3_omni.md
-  - General:
-    - usage/*
-  - Configuration:
-    - configuration/README.md
-    - configuration/*
-  - Diffusion Acceleration:
-     - TeaCache: user_guide/teacache.md
-  - Models:
-    - models/supported_models.md
-- Developer Guide:
-  - General:
-    - contributing/README.md
-    - glob: contributing/*
-      flatten_single_child_sections: true
-  - Model Implementation:
-    - contributing/model/README.md
-  - CI: contributing/ci
-  - Design Documents:
-    - design/index.md
-    - design/architecture_overview.md
-    - Feature Design:
-      - design/feature/multi_request_streaming_design.md
-      - design/feature/omni_connector_design.md
-      - design/feature/ray_based_execution.md
-    - Module Design:
-      - design/module/ar_module_design.md
-      - design/module/dit_module_design.md
-      - design/module/entrypoint_module_design.md
-  - Docs Guide: contributing/DOCS_GUIDE.md
-- API Reference:
-  - api/README.md
-  - api/vllm_omni
-- CLI Reference: cli
-- Community:
-  - community/*
-  - Slack: https://slack.vllm.ai
-  - Blog: https://blog.vllm.ai
-  - Forum: https://discuss.vllm.ai

docs/DOCS_GUIDE.md

@@ -1,6 +1,6 @@
 # Documentation Build Guide
 
-This directory contains the source files for the vLLM-Omni documentation.
+This directory contains the source files for the vLLM-omni documentation.
 
 ## Building Documentation Locally
 
@@ -9,7 +9,7 @@ This directory contains the source files for the vLLM-Omni documentation.
 Install documentation dependencies:
 
 ```bash
-uv pip install -e ".[docs]"
+pip install -e ".[docs]"
 ```
 
 ### Build and Serve Documentation
@@ -38,8 +38,8 @@ The documentation automatically extracts docstrings from the code using mkdocstr
 Example docstring:
 
 ```python
-class Omni:
-    """Main entry point for vLLM-Omni inference.
+class OmniLLM:
+    """Main entry point for vLLM-omni inference.
 
     This class provides a high-level interface for running multi-modal
     inference with non-autoregressive models.
@@ -50,7 +50,7 @@ class Omni:
         **kwargs: Additional arguments passed to the engine
 
     Example:
-        >>> llm = Omni(model="Qwen/Qwen2.5-Omni")
+        >>> llm = OmniLLM(model="Qwen/Qwen2.5-Omni")
         >>> outputs = llm.generate(prompts="Hello")
     """
 ```
@@ -84,7 +84,7 @@ The documentation is automatically deployed to GitHub Pages using GitHub Actions
    ```
 
 3. **Documentation will be available at**:
-   - `https://vllm-omni.readthedocs.io`
+   - `https://vllm-project.github.io/vllm-omni/`
 
 The GitHub Actions workflow (`.github/workflows/docs.yml`) will automatically:
 - Build the documentation when you push to `main` branch

docs/README.md

@@ -1,64 +0,0 @@
----
-hide:
-  - navigation
-  - toc
----
-
-# Welcome to vLLM-Omni
-
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="./source/logos/vllm-omni-logo.png">
-    <img alt="vllm-omni" src="./source/logos/vllm-omni-logo.png" width=55%>
-  </picture>
-</p>
-<h3 align="center">
-Easy, fast, and cheap omni-modality model serving for everyone
-</h3>
-
-<p style="text-align:center">
-<script async defer src="https://buttons.github.io/buttons.js"></script>
-<a class="github-button" href="https://github.com/vllm-project/vllm-omni" data-show-count="true" data-size="large" aria-label="Star">Star</a>
-<a class="github-button" href="https://github.com/vllm-project/vllm-omni/subscription" data-show-count="true" data-icon="octicon-eye" data-size="large" aria-label="Watch">Watch</a>
-<a class="github-button" href="https://github.com/vllm-project/vllm-omni/fork" data-show-count="true" data-icon="octicon-repo-forked" data-size="large" aria-label="Fork">Fork</a>
-</p>
-
-
-## About
-
-[vLLM](https://github.com/vllm-project/vllm) was originally designed to support large language models for text-based autoregressive generation tasks. vLLM-Omni is a framework that extends its support for omni-modality model inference and serving:
-
-- **Omni-modality**: Text, image, video, and audio data processing
-- **Non-autoregressive Architectures**: extend the AR support of vLLM to Diffusion Transformers (DiT) and other parallel generation models
-- **Heterogeneous outputs**: from traditional text generation to multimodal outputs
-
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="./source/architecture/omni-modality-model-architecture.png">
-    <img alt="vllm-omni-arch" src="./source/architecture/omni-modality-model-architecture.png" width=55%>
-  </picture>
-</p>
-
-vLLM-Omni is fast with:
-
-- State-of-the-art AR support by leveraging efficient KV cache management from vLLM
-- Pipelined stage execution overlapping for high throughput performance
-- Fully disaggregation based on OmniConnector and dynamic resource allocation across stages
-
-vLLM-Omni is flexible and easy to use with:
-
-- Heterogeneous pipeline abstraction to manage complex model workflows
-- Seamless integration with popular Hugging Face models
-- Tensor, pipeline, data and expert parallelism support for distributed inference
-- Streaming outputs
-- OpenAI-compatible API server
-
-vLLM-Omni seamlessly supports most popular open-source models on HuggingFace, including:
-
-- Omni-modality models (e.g. Qwen2.5-Omni, Qwen3-Omni)
-- Multi-modality generation models (e.g. Qwen-Image)
-
-For more information, checkout the following:
-
-- [vllm-omni architecture design and recent roadmaps](https://docs.google.com/presentation/d/1Y7t2Zm3BIISPN-_X_sDpxZMNCWir2rAx/edit?usp=drive_link&ouid=110473603432222024453&rtpof=true&sd=true)
-- [vllm-omni announcement blogpost](https://blog.vllm.ai/2025/11/30/vllm-omni.html)

docs/api/README.md

@@ -1,118 +0,0 @@
-# Summary
-
-## Entry Points
-
-Main entry points for vLLM-Omni inference and serving.
-
-- [vllm_omni.entrypoints.async_omni.AsyncOmni][]
-- [vllm_omni.entrypoints.async_omni.AsyncOmniStageLLM][]
-- [vllm_omni.entrypoints.chat_utils.OmniAsyncMultiModalContentParser][]
-- [vllm_omni.entrypoints.chat_utils.OmniAsyncMultiModalItemTracker][]
-- [vllm_omni.entrypoints.chat_utils.parse_chat_messages_futures][]
-- [vllm_omni.entrypoints.cli.serve.OmniServeCommand][]
-- [vllm_omni.entrypoints.log_utils.OrchestratorMetrics][]
-- [vllm_omni.entrypoints.omni.Omni][]
-- [vllm_omni.entrypoints.omni_llm.OmniLLM][]
-- [vllm_omni.entrypoints.omni_llm.OmniStageLLM][]
-- [vllm_omni.entrypoints.omni_stage.OmniStage][]
-- [vllm_omni.entrypoints.openai.serving_chat.OmniOpenAIServingChat][]
-
-## Inputs
-
-Input data structures for multi-modal inputs.
-
-- [vllm_omni.inputs.data.OmniEmbedsPrompt][]
-- [vllm_omni.inputs.data.OmniTokenInputs][]
-- [vllm_omni.inputs.data.OmniTokensPrompt][]
-- [vllm_omni.inputs.parse.parse_singleton_prompt_omni][]
-- [vllm_omni.inputs.preprocess.OmniInputPreprocessor][]
-
-## Engine
-
-Engine classes for offline and online inference.
-
-- [vllm_omni.diffusion.diffusion_engine.DiffusionEngine][]
-- [vllm_omni.engine.AdditionalInformationEntry][]
-- [vllm_omni.engine.AdditionalInformationPayload][]
-- [vllm_omni.engine.OmniEngineCoreOutput][]
-- [vllm_omni.engine.OmniEngineCoreOutputs][]
-- [vllm_omni.engine.OmniEngineCoreRequest][]
-- [vllm_omni.engine.PromptEmbedsPayload][]
-- [vllm_omni.engine.arg_utils.AsyncOmniEngineArgs][]
-- [vllm_omni.engine.arg_utils.OmniEngineArgs][]
-- [vllm_omni.engine.output_processor.MultimodalOutputProcessor][]
-- [vllm_omni.engine.output_processor.OmniRequestState][]
-- [vllm_omni.engine.processor.OmniProcessor][]
-
-## Core
-
-Core scheduling and caching components.
-
-- [vllm_omni.core.dit_cache_manager.DiTCacheManager][]
-- [vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler][]
-- [vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler][]
-- [vllm_omni.core.sched.output.OmniNewRequestData][]
-
-## Model Executor
-
-Model execution components.
-
-- [vllm_omni.model_executor.models.output_templates.OmniOutput][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni.Qwen2_5OmniForConditionalGeneration][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_talker.Qwen2_5OmniTalkerForConditionalGeneration][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_thinker.Qwen2_5OmniAudioFeatureInputs][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_thinker.Qwen2_5OmniConditionalGenerationMixin][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_thinker.Qwen2_5OmniThinkerDummyInputsBuilder][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_thinker.Qwen2_5OmniThinkerForConditionalGeneration][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_thinker.Qwen2_5OmniThinkerMultiModalDataParser][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_thinker.Qwen2_5OmniThinkerMultiModalProcessor][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_thinker.Qwen2_5OmniThinkerProcessingInfo][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_token2wav.Qwen2_5OmniToken2WavBigVGANModel][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_token2wav.Qwen2_5OmniToken2WavDiTModel][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_token2wav.Qwen2_5OmniToken2WavForConditionalGenerationVLLM][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_5_omni_token2wav.Qwen2_5OmniToken2WavModel][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_old.Qwen2EmbeddingModel][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_old.Qwen2ForCausalLM][]
-- [vllm_omni.model_executor.models.qwen2_5_omni.qwen2_old.Qwen2Model][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_moe.Qwen3MoeForCausalLM][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni.Qwen3OmniMoeForConditionalGeneration][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_code2wav.Qwen3OmniMoeCode2Wav][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_code_predictor_mtp.Qwen3OmniCodePredictorBaseModel][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_code_predictor_mtp.Qwen3OmniMoeTalkerCodePredictor][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_talker.Qwen3OmniMoeModel][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_talker.Qwen3OmniMoeTalkerForConditionalGeneration][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3MoeLLMForCausalLM][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3MoeLLMModel][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3OmniMoeConditionalGenerationMixin][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3OmniMoeThinkerForConditionalGeneration][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3OmniMoeThinkerMultiModalProcessor][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3OmniMoeThinkerProcessingInfo][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3Omni_VisionTransformer][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3_VisionPatchEmbed][]
-- [vllm_omni.model_executor.models.qwen3_omni.qwen3_omni_moe_thinker.Qwen3_VisionPatchMerger][]
-
-## Configuration
-
-Configuration classes.
-
-- [vllm_omni.config.model.OmniModelConfig][]
-- [vllm_omni.diffusion.cache.teacache.config.TeaCacheConfig][]
-
-## Workers
-
-Worker classes and model runners for distributed inference.
-
-- [vllm_omni.diffusion.worker.gpu_worker.GPUWorker][]
-- [vllm_omni.diffusion.worker.gpu_worker.WorkerProc][]
-- [vllm_omni.diffusion.worker.npu.npu_worker.NPUWorker][]
-- [vllm_omni.diffusion.worker.npu.npu_worker.NPUWorkerProc][]
-- [vllm_omni.worker.gpu_ar_model_runner.GPUARModelRunner][]
-- [vllm_omni.worker.gpu_ar_worker.GPUARWorker][]
-- [vllm_omni.worker.gpu_generation_model_runner.GPUGenerationModelRunner][]
-- [vllm_omni.worker.gpu_generation_worker.GPUGenerationWorker][]
-- [vllm_omni.worker.gpu_model_runner.OmniGPUModelRunner][]
-- [vllm_omni.worker.npu.npu_ar_model_runner.NPUARModelRunner][]
-- [vllm_omni.worker.npu.npu_ar_worker.NPUARWorker][]
-- [vllm_omni.worker.npu.npu_generation_model_runner.NPUGenerationModelRunner][]
-- [vllm_omni.worker.npu.npu_generation_worker.NPUGenerationWorker][]
-- [vllm_omni.worker.npu.npu_model_runner.OmniNPUModelRunner][]

docs/api/overview.md

@@ -0,0 +1,68 @@
+# Summary
+
+## Entry Points
+
+Main entry points for vLLM-omni inference and serving.
+
+- [vllm_omni.entrypoints.omni_llm.OmniLLM][]
+- [vllm_omni.entrypoints.omni_llm.OmniStageLLM][]
+- [vllm_omni.entrypoints.omni_stage.OmniStage][]
+- [vllm_omni.entrypoints.cli.serve.OmniServeCommand][]
+
+## Inputs
+
+Input data structures for multi-modal inputs.
+
+- [vllm_omni.inputs.data.OmniTokenInputs][]
+- [vllm_omni.inputs.data.OmniTokensPrompt][]
+- [vllm_omni.inputs.data.OmniEmbedsPrompt][]
+- [vllm_omni.inputs.parse.parse_singleton_prompt_omni][]
+- [vllm_omni.inputs.preprocess.OmniInputPreprocessor][]
+
+## Engine
+
+Engine classes for offline and online inference.
+
+- [vllm_omni.engine.arg_utils.OmniEngineArgs][]
+- [vllm_omni.engine.processor.OmniProcessor][]
+- [vllm_omni.engine.output_processor.OmniRequestState][]
+- [vllm_omni.engine.output_processor.MultimodalOutputProcessor][]
+- [vllm_omni.engine.PromptEmbedsPayload][]
+- [vllm_omni.engine.AdditionalInformationPayload][]
+- [vllm_omni.engine.OmniEngineCoreRequest][]
+
+## Core
+
+Core scheduling and caching components.
+
+- [vllm_omni.core.sched.scheduler.OmniScheduler][]
+- [vllm_omni.core.sched.diffusion_scheduler.DiffusionScheduler][]
+- [vllm_omni.core.sched.output.OmniNewRequestData][]
+- [vllm_omni.core.dit_cache_manager.DiTCacheManager][]
+
+## Model Executor
+
+Model execution components.
+
+- [vllm_omni.model_executor.models.qwen2_5_omni.Qwen2_5OmniForConditionalGeneration][]
+- [vllm_omni.model_executor.models.qwen2_5_omni_talker.Qwen2_5OmniTalkerForConditionalGeneration][]
+- [vllm_omni.model_executor.models.qwen2_5_omni_thinker.Qwen2_5OmniThinkerForConditionalGeneration][]
+- [vllm_omni.model_executor.models.qwen2_5_omni_token2wav.Qwen2_5OmniToken2WavForConditionalGenerationVLLM][]
+- [vllm_omni.model_executor.layers.mrope.MRotaryEmbedding][]
+- [vllm_omni.model_executor.models.registry.OmniModelRegistry][]
+
+## Configuration
+
+Configuration classes.
+
+- [vllm_omni.config.OmniModelConfig][]
+
+## Workers
+
+Worker classes and model runners for distributed inference.
+
+- [vllm_omni.worker.gpu_ar_worker.GPUARWorker][]
+- [vllm_omni.worker.gpu_diffusion_worker.GPUDiffusionWorker][]
+- [vllm_omni.worker.gpu_model_runner.OmniGPUModelRunner][]
+- [vllm_omni.worker.gpu_ar_model_runner.GPUARModelRunner][]
+- [vllm_omni.worker.gpu_diffusion_model_runner.GPUDiffusionModelRunner][]

docs/cli/README.md

@@ -1,24 +0,0 @@
-# vLLM-Omni CLI Guide
-
-The CLI for vLLM-Omni inherits from vllm with some additional arguments.
-
-## serve
-
-Starts the vLLM-Omni OpenAI Compatible API server.
-
-Start with a model:
-
-```bash
-vllm serve Qwen/Qwen2.5-Omni-7B --omni
-```
-
-Specify the port:
-
-```bash
-vllm serve Qwen/Qwen2.5-Omni-7B --omni --port 8091
-```
-
-If you have custom stage configs file, launch the server with command below
-```bash
-vllm serve Qwen/Qwen2.5-Omni-7B --omni --stage-configs-path /path/to/stage_configs_file
-```

docs/community/contact_us.md

@@ -1,5 +0,0 @@
-# Contact Us
-
-- 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

docs/community/meetups.md

@@ -1 +0,0 @@
-# Meetups

docs/community/volunteers.md

@@ -1,12 +0,0 @@
-# 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!

docs/configuration/README.md

@@ -1,15 +0,0 @@
-# Configuration Options
-
-This section lists the most common options for running vLLM-Omni.
-
-For options within a vLLM Engine. Please refer to [vLLM Configuration](https://docs.vllm.ai/en/v0.11.0/configuration/index.html)
-
-Currently, the main options are maintained by stage configs for each model.
-
-For specific example, please refer to [Qwen2.5-omni stage config](stage_configs/qwen2_5_omni.yaml)
-
-For introduction, please check [Introduction for stage config](./stage_configs.md)
-
-## Optimization Features
-
-- **[TeaCache Configuration](../user_guide/teacache.md)** - Enable TeaCache adaptive caching for DiT models to achieve 1.5x-2.0x speedup with minimal quality loss

docs/configuration/stage_configs.md

@@ -1,272 +0,0 @@
-# Stage configs for vLLM-Omni
-
-In vLLM-Omni, the target model is separated into multiple stages, which are processed by different LLMEngines, DiffusionEngines or other types of engines. Depending on different types of stages, such as Autoregressive (AR) stage or Diffusion transformer (DiT) stage, each can choose corresponding schedulers, model workers to load with the Engines in a plug-in fashion.
-
-!!! note
-    Default stage config YAMLs (for example, `vllm_omni/model_executor/stage_configs/qwen2_5_omni.yaml` and `vllm_omni/model_executor/stage_configs/qwen3_omni_moe.yaml`) are bundled and loaded automatically when `stage_configs_path` is not provided. They have been verified to work on 1xH100 for Qwen2.5-Omni and 2xH100 for Qwen3-Omni.
-
-Therefore, as a core part of vLLM-Omni, the stage configs for a model have several main functions:
-
-- Claim partition of stages and their corresponding class implementation in `model_executor/models`.
-- The disaggregated configuration for each stage and the communication topology among them.
-- Engine arguments for each engine within the stage.
-- Input and output dependencies for each stage.
-- Default input parameters.
-
-If users want to modify some part of it. The custom stage_configs file can be input as input argument in both online and offline. Just like examples below:
-
-For offline (Assume necessary dependencies have ben imported):
-```python
-model_name = "Qwen/Qwen2.5-Omni-7B"
-omni_llm = OmniLLM(model=model_name, stage_configs_path="/path/to/custom_stage_configs.yaml")
-```
-
-For online serving:
-```bash
-vllm serve Qwen/Qwen2.5-Omni-7B --omni --port 8091 --stage-configs-path /path/to/stage_configs_file
-```
-!!! important
-    We are actively iterating on the definition of stage configs, and we welcome all feedbacks from both community users and developers to help us shape the development!
-
-Below is a specific example of stage_configs.yaml in Qwen2.5-omni.
-```python
-# stage config for running qwen2.5-omni with architecture of OmniLLM.
-stage_args:
-  - stage_id: 0 # mark the unique id for each stage
-    runtime: # The disaggregated configuration
-      process: true  # Run this stage in a separate process
-      devices: "0" # Visible devices for this stage (CUDA_VISIBLE_DEVICES/torch.cuda.set_device)
-      max_batch_size: 1 # the batch_size for offline inference
-    engine_args: # Engine arguments for a certain engine
-      model_stage: thinker
-      model_arch: Qwen2_5OmniForConditionalGeneration # The model implementation registered in model_executor/models/registry.py
-      worker_cls: vllm_omni.worker.gpu_ar_worker.GPUARWorker # The specific worker used
-      scheduler_cls: vllm_omni.core.sched.scheduler.OmniScheduler # The specific scehduler used
-      gpu_memory_utilization: 0.8 # The gpu memory allocation for the stage within a single chip
-      enforce_eager: true  # Now we only support eager mode
-      trust_remote_code: true # Needed by huggingface config parsing
-      engine_output_type: latent  # It claims that the stage will input latent hiddenstates besides token ids
-      enable_prefix_caching: false # For request with hiddenstates output, the prefix caching is not supported now
-    is_comprehension: true # If the stage is a text or multimodal comprehension module. If it is, the AsyncOmni will use its tokenizer as default
-    final_output: true # If the stage has output as part of final outputs. If it is false, which means that the stage only works as a intermediate role.
-    final_output_type: text # What is the final output type. It can be text and audio now.
-    default_sampling_params: # sampling parameters for the stage. Their meaning aligns with vLLM.
-      temperature: 0.0
-      top_p: 1.0
-      top_k: -1
-      max_tokens: 2048
-      seed: 42
-      detokenize: True
-      repetition_penalty: 1.1
-  - stage_id: 1
-    runtime:
-      process: true
-      devices: "1"
-      max_batch_size: 3
-    engine_args:
-      model_stage: talker
-      model_arch: Qwen2_5OmniForConditionalGeneration
-      worker_cls: vllm_omni.worker.gpu_ar_worker.GPUARWorker
-      scheduler_cls: vllm_omni.core.sched.scheduler.OmniScheduler
-      gpu_memory_utilization: 0.8
-      enforce_eager: true
-      trust_remote_code: true
-      enable_prefix_caching: false
-      engine_output_type: latent
-    engine_input_source: [0]
-    custom_process_input_func: vllm_omni.model_executor.stage_input_processors.qwen2_5_omni.thinker2talker
-    default_sampling_params:
-      temperature: 0.9
-      top_p: 0.8
-      top_k: 40
-      max_tokens: 2048
-      seed: 42
-      detokenize: True
-      repetition_penalty: 1.05
-      stop_token_ids: [8294]
-  - stage_id: 2
-    runtime:
-      process: true
-      devices: "0"            # Example: use a different GPU than the previous stage; use "0" if single GPU
-      max_batch_size: 1
-    engine_args:
-      model_stage: code2wav
-      model_arch: Qwen2_5OmniForConditionalGeneration
-      worker_cls: vllm_omni.worker.gpu_diffusion_worker.GPUDiffusionWorker
-      scheduler_cls: vllm_omni.core.sched.diffusion_scheduler.DiffusionScheduler
-      gpu_memory_utilization: 0.15
-      enforce_eager: true
-      trust_remote_code: true
-      enable_prefix_caching: false
-      engine_output_type: audio
-    engine_input_source: [1]
-    final_output: true
-    final_output_type: audio
-    default_sampling_params:
-      temperature: 0.0
-      top_p: 1.0
-      top_k: -1
-      max_tokens: 2048
-      seed: 42
-      detokenize: True
-      repetition_penalty: 1.1
-
-# Top-level runtime config (concise): default windows and stage edges
-runtime:
-  enabled: true
-  defaults:
-    window_size: -1             # Simplified: trigger downstream only after full upstream completion
-    max_inflight: 1             # Simplified: process serially within each stage
-  edges:
-    - from: 0                   # thinker → talker: trigger only after receiving full input (-1)
-      to: 1
-      window_size: -1
-    - from: 1                   # talker → code2wav: trigger only after receiving full input (-1)
-      to: 2
-      window_size: -1
-
-```
-
-## Stage Configuration Arguments
-
-Each stage in the `stage_args` list contains the following configuration options:
-
-### `stage_id`
-
-A unique identifier for each stage in the multi-stage pipeline. Stages are numbered sequentially starting from 0, and this ID is used to reference stages in inter-stage dependencies (e.g., `engine_input_source`).
-
-### `runtime`
-
-Configuration for disaggregated execution of the stage, controlling how the stage is deployed and executed.
-
-#### `runtime.process`
-
-Whether to run this stage in a separate process. When set to `true`, the stage will be executed in an isolated process, enabling better resource isolation and parallel execution across different stages. This is essential for multi-GPU deployments where different stages run on different devices.
-
-Default: `true`
-
-#### `runtime.devices`
-
-Visible devices for this stage, specified as a string. This controls which GPU devices are available to the stage process, similar to setting `CUDA_VISIBLE_DEVICES` or using `torch.cuda.set_device()`. For example, `"0"` uses GPU 0, `"1"` uses GPU 1, and `"0,1"` makes both GPUs 0 and 1 visible.
-
-Default: `"0"`
-
-#### `runtime.max_batch_size`
-
-The maximum batch size for offline inference in this stage. This limits how many sequences can be processed together in a single batch during offline inference operations.
-
-Default: `1`
-
-### `engine_args`
-
-Engine arguments for configuring the LLM engine, diffusion engine, or other engine types used by this stage.
-
-#### `engine_args.model_stage`
-
-The name identifier for this model stage within the multi-stage architecture. This is used internally to distinguish different stages of the same model (e.g., "thinker", "talker", "code2wav" in Qwen2.5-Omni).
-
-#### `engine_args.model_arch`
-
-The model architecture class name that is registered in `model_executor/models/registry.py`. This specifies which model implementation to use for this stage. The class must be registered in the model registry for vLLM-Omni to locate and instantiate it.
-
-#### `engine_args.worker_cls`
-
-The specific worker class to use for this stage. This determines how the model computations are executed. Examples include `vllm_omni.worker.gpu_ar_worker.GPUARWorker` for autoregressive stages and `vllm_omni.worker.gpu_diffusion_worker.GPUDiffusionWorker` for diffusion-based stages.
-
-#### `engine_args.scheduler_cls`
-
-The scheduler class to use for this stage. The scheduler manages request queuing, batching, and execution order. Examples include `vllm_omni.core.sched.scheduler.OmniScheduler` for standard stages and `vllm_omni.core.sched.diffusion_scheduler.DiffusionScheduler` for diffusion stages.
-
-#### `engine_args.gpu_memory_utilization`
-
-The fraction of GPU memory to allocate for this stage within a single GPU chip. This is a value between 0.0 and 1.0, where 0.8 means 80% of the GPU memory will be used by this stage. This allows fine-grained control over memory allocation when multiple stages share the same GPU or when reserving memory for other operations.
-
-Default: `0.8`
-
-#### `engine_args.enforce_eager`
-
-Whether to enforce eager execution mode. When set to `true`, the engine will run in eager mode without using CUDA graphs or other compilation optimizations. Currently, vLLM-Omni only supports eager mode.
-
-Default: `true`
-
-#### `engine_args.trust_remote_code`
-
-Whether to trust remote code when loading models from Hugging Face. This is required for models that use custom code in their configuration files. Set to `true` when loading models that require custom model implementations.
-
-Default: `true`
-
-#### `engine_args.engine_output_type`
-
-Specifies the type of output produced by this stage's engine. This determines what kind of data flows to downstream stages. Possible values include `latent` (hidden states), `text` (tokenized text), and `audio` (audio waveforms). When set to `latent`, the stage outputs latent hidden states in addition to token IDs, which are consumed by downstream stages.
-
-Default: `latent`
-
-#### `engine_args.enable_prefix_caching`
-
-Whether to enable prefix caching for this stage. Prefix caching can improve performance by caching KV cache for common prompt prefixes. However, for requests that output hidden states (when `engine_output_type` is `latent`), prefix caching is not currently supported and should be set to `false`.
-
-Default: `false`
-
-### `is_comprehension`
-
-Whether this stage is a text or multimodal comprehension module. When set to `true`, the stage acts as a comprehension module that processes input text or multimodal content. If this is the first comprehension stage, `AsyncOmni` will use its tokenizer as the default tokenizer for the entire pipeline.
-
-Default: `true`
-
-### `final_output`
-
-Whether this stage produces output that is part of the final outputs returned to the user. When set to `false`, the stage only works as an intermediate stage, processing data that flows to downstream stages but not contributing directly to the final response.
-
-Default: `true`
-
-### `final_output_type`
-
-The type of final output produced by this stage. This specifies what format the output will be in when returned to the user. Currently supported values are `text` (for text generation) and `audio` (for audio generation).
-
-Default: `text`
-
-### `default_sampling_params`
-
-Default sampling parameters for this stage. These parameters control the generation behavior and align with vLLM's sampling parameter semantics. These defaults are used when no explicit sampling parameters are provided in the request.
-
-#### `default_sampling_params.temperature`
-
-Sampling temperature for controlling randomness. Lower values (e.g., 0.0) make the output more deterministic and focused, while higher values increase randomness.
-
-Default: `0.0`
-
-#### `default_sampling_params.top_p`
-
-Nucleus sampling parameter. Only tokens with cumulative probability mass up to `top_p` are considered. This helps filter out low-probability tokens while maintaining diversity.
-
-Default: `1.0`
-
-#### `default_sampling_params.top_k`
-
-Top-k sampling parameter. Only the top `k` most likely tokens are considered. Set to `-1` to disable top-k filtering and consider all tokens.
-
-Default: `-1`
-
-#### `default_sampling_params.max_tokens`
-
-Maximum number of tokens to generate in this stage. This limits the length of the output sequence.
-
-Default: `2048`
-
-#### `default_sampling_params.seed`
-
-Random seed for reproducible generation. When set, the random number generator will be initialized with this seed to ensure consistent outputs across runs.
-
-Default: `42`
-
-#### `default_sampling_params.detokenize`
-
-Whether to detokenize the output tokens into text. When set to `true`, token IDs are converted back to readable text strings.
-
-Default: `True`
-
-#### `default_sampling_params.repetition_penalty`
-
-Penalty applied to tokens that have already appeared in the generated sequence. Values greater than 1.0 discourage repetition, while values less than 1.0 encourage it. A value of 1.0 applies no penalty.
-
-Default: `1.1`

docs/configuration/stage_configs/qwen2_5_omni.yaml

@@ -1,94 +0,0 @@
-# stage config for running qwen2.5-omni with architecture of OmniLLM.
-stage_args:
-  - stage_id: 0
-    runtime:
-      process: true            # Run this stage in a separate process
-      devices: "0"            # Visible devices for this stage (CUDA_VISIBLE_DEVICES/torch.cuda.set_device)
-      max_batch_size: 1
-    engine_args:
-      model_stage: thinker
-      model_arch: Qwen2_5OmniForConditionalGeneration
-      worker_cls: vllm_omni.worker.gpu_ar_worker.GPUARWorker
-      scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler
-      gpu_memory_utilization: 0.8
-      enforce_eager: true  # Now we only support eager mode
-      trust_remote_code: true
-      engine_output_type: latent
-      enable_prefix_caching: false
-    is_comprehension: true
-    final_output: true
-    final_output_type: text
-    default_sampling_params:
-      temperature: 0.0
-      top_p: 1.0
-      top_k: -1
-      max_tokens: 2048
-      seed: 42
-      detokenize: True
-      repetition_penalty: 1.1
-  - stage_id: 1
-    runtime:
-      process: true
-      devices: "1"
-      max_batch_size: 1
-    engine_args:
-      model_stage: talker
-      model_arch: Qwen2_5OmniForConditionalGeneration
-      worker_cls: vllm_omni.worker.gpu_ar_worker.GPUARWorker
-      scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler
-      gpu_memory_utilization: 0.8
-      enforce_eager: true
-      trust_remote_code: true
-      enable_prefix_caching: false
-      engine_output_type: latent
-    engine_input_source: [0]
-    custom_process_input_func: vllm_omni.model_executor.stage_input_processors.qwen2_5_omni.thinker2talker
-    default_sampling_params:
-      temperature: 0.9
-      top_p: 0.8
-      top_k: 40
-      max_tokens: 2048
-      seed: 42
-      detokenize: True
-      repetition_penalty: 1.05
-      stop_token_ids: [8294]
-  - stage_id: 2
-    runtime:
-      process: true
-      devices: "0"            # Example: use a different GPU than the previous stage; use "0" if single GPU
-      max_batch_size: 1
-    engine_args:
-      model_stage: code2wav
-      model_arch: Qwen2_5OmniForConditionalGeneration
-      worker_cls: vllm_omni.worker.gpu_diffusion_worker.GPUDiffusionWorker
-      scheduler_cls: vllm_omni.core.sched.diffusion_scheduler.DiffusionScheduler
-      gpu_memory_utilization: 0.15
-      enforce_eager: true
-      trust_remote_code: true
-      enable_prefix_caching: false
-      engine_output_type: audio
-    engine_input_source: [1]
-    final_output: true
-    final_output_type: audio
-    default_sampling_params:
-      temperature: 0.0
-      top_p: 1.0
-      top_k: -1
-      max_tokens: 2048
-      seed: 42
-      detokenize: True
-      repetition_penalty: 1.1
-
-# Top-level runtime config (concise): default windows and stage edges
-runtime:
-  enabled: true
-  defaults:
-    window_size: -1             # Simplified: trigger downstream only after full upstream completion
-    max_inflight: 1             # Simplified: process serially within each stage
-  edges:
-    - from: 0                   # thinker → talker: trigger only after receiving full input (-1)
-      to: 1
-      window_size: -1
-    - from: 1                   # talker → code2wav: trigger only after receiving full input (-1)
-      to: 2
-      window_size: -1

docs/contributing/README.md

@@ -1,52 +1,55 @@
-# Contributing to vLLM-Omni
+# Contributing to vLLM-omni
 
-Thank you for your interest in contributing to vLLM-Omni! This document provides guidelines and instructions for contributing.
-
-!!! note
-    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).
+Thank you for your interest in contributing to vLLM-omni! This document provides guidelines and instructions for contributing.
 
 ## Getting Started
 
-vLLM-Omni uses `uv` as the environment manager, to create and manage Python environments. Please follow the documentation to install `uv`. After installing `uv`, you can create a new Python environment using the following commands:
+### Development Environment
+
+vLLM-omni uses `uv` as the environment manager, to create and manage Python environments. Please follow the documentation to install `uv`. After installing `uv`, you can create a new Python environment using the following commands:
 
 ```bash
 uv venv --python 3.12 --seed
 source .venv/bin/activate
 ```
-### Development Environment for vLLM
 
-vLLM-Omni is built on top of the stable release 0.11.0 of vLLM. You may install `vllm==0.11.0` directly or install the library from source if you need to check, modify or debug with source code of vLLM.
+Install vLLM from source:
 
-### Development Environment for vLLM-Omni
+```bash
+git clone https://github.com/vllm-project/vllm.git
+cd vllm
+git checkout v0.11.0
+VLLM_USE_PRECOMPILED=1 uv pip install -e .
+```
 
-Install vLLM-Omni from source with development dependencies:
+Install vLLM-omni from source:
 
 ```bash
 git clone https://github.com/vllm-project/vllm-omni.git
-cd vllm-omni
-uv pip install -e ".[dev]"
+cd vllm_omni
+uv pip install -e .
 ```
 
 !!! tip
-    vLLM-Omni is compatible with Python versions 3.10 to 3.12. However, we recommend developing with Python 3.12 to minimize the chance of your local environment clashing with our CI environment.
+    vLLM-omni is compatible with Python versions 3.10 to 3.12. However, we recommend developing with Python 3.12 to minimize the chance of your local environment clashing with our CI environment.
 
 ### Linting
 
-vLLM-Omni uses `pre-commit` to lint and format the codebase. See [pre-commit documentation](https://pre-commit.com/#usage) if `pre-commit` is new to you. Setting up `pre-commit` is as easy as:
+vLLM-omni uses `pre-commit` to lint and format the codebase. See [pre-commit documentation](https://pre-commit.com/#usage) if `pre-commit` is new to you. Setting up `pre-commit` is as easy as:
 
 ```bash
 uv pip install pre-commit
 pre-commit install
 ```
 
-vLLM-Omni's `pre-commit` hooks will now run automatically every time you commit.
+vLLM-omni's `pre-commit` hooks will now run automatically every time you commit.
 
 !!! tip
     You can manually run the `pre-commit` hooks using:
 
     ```bash
     pre-commit run     # runs on staged files
-    pre-commit run --show-diff-on-failure --color=always --all-files  # runs on all files (short for --all-files)
+    pre-commit run -a  # runs on all files (short for --all-files)
     ```
 
 ### Documentation
@@ -75,14 +78,20 @@ For additional features and advanced configurations, refer to the:
 
 ### Testing
 
-vLLM-Omni uses `pytest` to test the codebase.
+vLLM-omni uses `pytest` to test the codebase.
 
 ```bash
+# Install the test dependencies used in CI (CUDA only)
+uv pip install -e ".[dev]" --torch-backend=auto
+
+# Install some common test dependencies (hardware agnostic)
+uv pip install pytest pytest-asyncio
+
 # Run all tests
 pytest tests/
 
 # Run tests for a single test file with detailed output
-pytest -s -v tests/test_omni_llm.py
+pytest -s -v tests/test_logger.py
 ```
 
 !!! warning
@@ -97,7 +106,7 @@ If you encounter a bug or have a feature request, please search existing issues
 
 ## Pull Requests & Code Reviews
 
-Thank you for your contribution to vLLM-Omni! Before submitting the pull request, please ensure the PR meets the following criteria. This helps vLLM-Omni maintain the code quality and improve the efficiency of the review process.
+Thank you for your contribution to vLLM-omni! Before submitting the pull request, please ensure the PR meets the following criteria. This helps vLLM-omni maintain the code quality and improve the efficiency of the review process.
 
 ### DCO and Signed-off-by
 
@@ -119,9 +128,9 @@ Only specific types of PRs will be reviewed. The PR title is prefixed appropriat
 - `[CI/Build]` for build or continuous integration improvements.
 - `[Doc]` for documentation fixes and improvements.
 - `[Model]` for adding a new model or improving an existing model. Model name should appear in the title.
-- `[Frontend]` For changes on the vLLM-Omni frontend (e.g., OpenAI API server, `OmniLLM` class, etc.)
+- `[Frontend]` For changes on the vLLM-omni frontend (e.g., OpenAI API server, `OmniLLM` class, etc.)
 - `[Kernel]` for changes affecting CUDA kernels or other compute kernels.
-- `[Core]` for changes in the core vLLM-Omni logic (e.g., `OmniProcessor`, `OmniScheduler`, etc.)
+- `[Core]` for changes in the core vLLM-omni logic (e.g., `OmniProcessor`, `OmniScheduler`, etc.)
 - `[Hardware][Vendor]` for hardware-specific changes. Vendor name should appear in the prefix, such as [Ascend] for Ascend NPUs.
 - `[Misc]` for PRs that do not fit the above categories. Please use this sparingly.
 
@@ -136,7 +145,7 @@ The PR needs to meet the following code quality standards:
 - Pass all linter checks.
 - The code needs to be well-documented to ensure future contributors can easily understand the code.
 - Include sufficient tests to ensure the project stays correct and robust. This includes both unit tests and integration tests.
-- Please add documentation to `docs/` if the PR modifies the user-facing behaviors of vLLM-Omni. It helps vLLM-Omni users understand and utilize the new features or changes.
+- Please add documentation to `docs/` if the PR modifies the user-facing behaviors of vLLM-omni. It helps vLLM-omni users understand and utilize the new features or changes.
 
 ### Notes for Large Changes
 
@@ -144,17 +153,18 @@ Please keep the changes as concise as possible. For major architectural changes
 
 ### What to Expect for the Reviews
 
-The goal of the vLLM-Omni team is to be a _transparent reviewing machine_. We would like to make the review process transparent and efficient and make sure no contributor feels confused or frustrated. However, the vLLM-Omni team is small, so we need to prioritize some PRs over others. Here is what you can expect from the review process:
+The goal of the vLLM-omni team is to be a _transparent reviewing machine_. We would like to make the review process transparent and efficient and make sure no contributor feels confused or frustrated. However, the vLLM-omni team is small, so we need to prioritize some PRs over others. Here is what you can expect from the review process:
 
 - After the PR is submitted, the PR will be assigned to a reviewer. Every reviewer will pick up the PRs based on their expertise and availability.
-- After the PR is assigned, the reviewer will provide status updates every 2-3 days. If the PR is not reviewed within 7 days, please feel free to ping the reviewer or the vLLM-Omni team.
+- After the PR is assigned, the reviewer will provide status updates every 2-3 days. If the PR is not reviewed within 7 days, please feel free to ping the reviewer or the vLLM-omni team.
 - After the review, the reviewer will put an `action-required` label on the PR if there are changes required. The contributor should address the comments and ping the reviewer to re-review the PR.
 - Please respond to all comments within a reasonable time frame. If a comment isn't clear or you disagree with a suggestion, feel free to ask for clarification or discuss the suggestion.
+- Note that not all CI checks will be executed due to limited computational resources. The reviewer will add `ready` label to the PR when the PR is ready to merge or a full CI run is needed.
 
 ## Additional Resources
 
-- [Design Documents](../design/index.md) - Architecture and design documentation
+- [Design Documents](design_documents/index.md) - Architecture and design documentation
 
 ## Thank You
 
-Finally, thank you for taking the time to read these guidelines and for your interest in contributing to vLLM-Omni. All of your contributions help make vLLM-Omni a great tool and community for everyone!
+Finally, thank you for taking the time to read these guidelines and for your interest in contributing to vLLM-omni. All of your contributions help make vLLM-omni a great tool and community for everyone!

docs/contributing/ci/failures.md

@@ -1,4 +0,0 @@
-# CI Failures
-
-What should I do when a CI job fails on my PR, but I don't think my PR caused
-the failure?

docs/contributing/design_documents/api_design_doc.md

@@ -0,0 +1,143 @@
+# API Design Documentation
+
+This document contains comprehensive API design documentation for all core modules in vLLM-omni. These templates provide a standardized structure for designing and implementing the core, engine, executor, and worker modules.
+
+## 📋 Module API
+
+### Core Module API
+**Core module** provides fundamental scheduling, caching, and resource management functionality.
+
+**Key Components:**
+- Request scheduling and prioritization
+- DiT cache management for diffusion models
+- Resource allocation and coordination
+- Inter-module communication
+
+### Engine Module API
+**Engine module** handles model loading, inference execution, and output processing.
+
+**Key Components:**
+- Model loading and initialization
+- Inference execution for AR and diffusion models
+- Input preprocessing for various modalities
+- Output postprocessing and formatting
+
+### Executor Module API
+**Executor module** coordinates and manages request execution across different engines and workers.
+
+**Key Components:**
+- Request routing and load balancing
+- Execution pipeline coordination
+- Worker management and task distribution
+- Error handling and recovery
+
+### Worker Module API
+**Worker module** provides the actual execution environment for model inference.
+
+**Key Components:**
+- Model execution and inference
+- GPU resource management
+- Request batching and processing
+- Performance optimization
+
+
+## 🔧 Implementation Guidelines
+
+### Code Standards
+- Use type hints for all methods
+- Include comprehensive docstrings
+- Follow PEP 8 style guidelines
+- Use async/await for I/O operations
+
+### Error Handling
+- Define specific exception types
+- Provide meaningful error messages
+- Include error recovery strategies
+- Log errors appropriately
+
+### Configuration
+- Use dataclasses for configuration
+- Provide sensible defaults
+- Include validation methods
+- Support environment variables
+
+### Testing
+- Write unit tests for all public methods
+- Include integration tests
+- Test error conditions
+- Validate configuration options
+
+## 🚀 Getting Started
+
+### For Developers
+
+1. **Choose a Module**
+   - Pick the module you want to work on
+   - Read the corresponding API documentation
+   - Understand the responsibilities and interfaces
+
+2. **Set Up Development Environment**
+   ```bash
+   # Clone the repository
+   git clone https://github.com/vllm-project/vllm-omni.git
+   cd vllm-omni
+
+   # Install dependencies
+   pip install -e ".[dev]"
+
+   # Set up pre-commit hooks
+   pre-commit install
+   ```
+
+3. **Start Implementation**
+   - Create the module directory
+   - Implement base classes first
+   - Add concrete implementations
+   - Write tests as you go
+
+### For Contributors
+
+1. **Review API Documentation**
+   - Read through all module APIs
+   - Understand the overall architecture
+   - Identify areas for improvement
+
+2. **Propose Changes**
+   - Create issues for API changes
+   - Discuss in pull requests
+   - Update documentation accordingly
+
+3. **Submit Contributions**
+   - Follow the coding standards
+   - Include tests for new features
+   - Update documentation
+   - Submit pull requests
+
+## 🤝 Contributing
+
+We welcome contributions to improve the API documentation and implementation. Please:
+
+1. **Follow the Template Structure**
+   - Use the [api template](https://github.com/vllm-project/vllm-omni/docs/contributing/design_documents/api_design_template.md) as a guide
+   - Maintain consistency across modules
+   - Include all required sections
+
+2. **Provide Working Examples**
+   - Test all code examples
+   - Include both basic and advanced usage
+   - Show integration patterns
+
+3. **Keep Documentation Updated**
+   - Update docs when APIs change
+   - Version control documentation changes
+   - Maintain backward compatibility notes
+
+## 📝 Notes
+
+- All API documentation should be kept up-to-date with implementation
+- Code examples should be tested and working
+- Configuration options should be validated
+- Error handling should be comprehensive
+- Performance considerations should be documented
+
+For questions or suggestions about the API documentation, please open an issue or start a discussion in the repository.

docs/contributing/design_documents/api_design_template.md

@@ -0,0 +1,336 @@
+# API Design Template for vLLM-omni Modules
+
+This template provides a standardized structure for designing APIs for core, engine, executor, and worker modules in vLLM-omni. Use this template to ensure consistency and completeness across all modules.
+
+## Template Structure
+
+### 1. Module Overview
+- **Purpose**: What this module does
+- **Responsibilities**: Key responsibilities of the module
+- **Dependencies**: What other modules this depends on
+- **Integration Points**: How it integrates with other modules
+
+### 2. Core Classes/Interfaces
+- **Base Classes**: Abstract base classes and interfaces
+- **Implementation Classes**: Concrete implementations
+- **Data Structures**: Key data structures and models
+
+### 3. Public API Methods
+- **Initialization**: Constructor and setup methods
+- **Core Operations**: Main functionality methods
+- **Configuration**: Configuration and parameter methods
+- **Lifecycle Management**: Start, stop, cleanup methods
+- **Monitoring**: Status, metrics, and debugging methods
+
+### 4. Configuration
+- **Configuration Classes**: Dataclasses or config objects
+- **Required Parameters**: Must-have configuration
+- **Optional Parameters**: Optional configuration with defaults
+- **Validation**: Parameter validation rules
+
+### 5. Error Handling
+- **Custom Exceptions**: Module-specific exceptions
+- **Error Codes**: Standardized error codes
+- **Recovery Strategies**: How to handle and recover from errors
+
+### 6. Examples
+- **Basic Usage**: Simple usage examples
+- **Advanced Usage**: Complex scenarios
+- **Integration Examples**: How to use with other modules
+
+---
+
+## Example: Core Module Template
+
+### 1. Module Overview
+
+**Purpose**: The core module provides fundamental scheduling and caching functionality for vLLM-omni.
+
+**Responsibilities**:
+- Request scheduling and prioritization
+- DiT cache management
+- Resource allocation and coordination
+- Inter-module communication
+
+**Dependencies**:
+- `vllm_omni.request` - Request handling
+- `vllm_omni.config` - Configuration management
+- `vllm_omni.utils` - Utility functions
+
+**Integration Points**:
+- Receives requests from entrypoints
+- Coordinates with engine modules
+- Manages worker allocation
+- Provides status to monitoring systems
+
+### 2. Core Classes/Interfaces
+
+```python
+from abc import ABC, abstractmethod
+from typing import List, Optional, Dict, Any
+from dataclasses import dataclass
+from enum import Enum
+
+class DiTCacheManager:
+    """Manages DiT cache for diffusion models."""
+
+    def __init__(self, config: DiTCacheConfig):
+        self.config = config
+        self.cache = {}
+        self.cache_stats = {}
+
+    async def get_cache(self, cache_key: str) -> Optional[Any]:
+        """Retrieve cached data."""
+        pass
+
+    async def set_cache(self, cache_key: str, data: Any) -> None:
+        """Store data in cache."""
+        pass
+
+    async def invalidate_cache(self, cache_key: str) -> None:
+        """Invalidate cached data."""
+        pass
+```
+
+### 3. Public API Methods
+
+#### Initialization
+```python
+class OmniScheduler():
+    def __init__(self, config: SchedulerConfig):
+        """Initialize the core scheduler."""
+        self.config = config
+        self.scheduler = self._create_scheduler()
+        self.cache_manager = DiTCacheManager(config.dit_cache_config)
+        self._running = False
+
+    def _create_scheduler(self) -> BaseScheduler:
+        """Factory method to create appropriate scheduler."""
+        if self.config.scheduler_type == SchedulerType.FIFO:
+            return FIFOScheduler(self.config)
+        elif self.config.scheduler_type == SchedulerType.PRIORITY:
+            return PriorityScheduler(self.config)
+        # ... other scheduler types
+```
+
+#### Core Operations
+```python
+    async def schedule(self, request: OmniRequest) -> bool:
+        """
+        Schedule a request for processing.
+
+        Args:
+            request: The request to schedule
+
+        Returns:
+            bool: True if successfully scheduled, False otherwise
+
+        Raises:
+            SchedulerError: If scheduling fails
+            QueueFullError: If queue is at capacity
+        """
+        pass
+```
+
+#### Configuration
+```python
+    def update_config(self, new_config: SchedulerConfig) -> None:
+        """Update scheduler configuration."""
+        pass
+
+    def get_config(self) -> SchedulerConfig:
+        """Get current configuration."""
+        pass
+```
+
+#### Lifecycle Management
+```python
+    async def start(self) -> None:
+        """Start the scheduler."""
+        self._running = True
+        # Start background tasks
+
+    async def stop(self) -> None:
+        """Stop the scheduler gracefully."""
+        self._running = False
+        # Cleanup resources
+
+    async def shutdown(self) -> None:
+        """Force shutdown the scheduler."""
+        # Immediate cleanup
+```
+
+#### Monitoring
+```python
+    def get_status(self) -> Dict[str, Any]:
+        """Get current scheduler status."""
+        return {
+            "running": self._running,
+            "queue_size": self.scheduler.queue_size(),
+            "processed_requests": self.scheduler.processed_count(),
+            "cache_hit_rate": self.cache_manager.get_hit_rate()
+        }
+
+    def get_metrics(self) -> Dict[str, float]:
+        """Get performance metrics."""
+        pass
+```
+
+### 4. Configuration
+
+```python
+@dataclass
+class OmniConfig:
+    """Configuration for the core module."""
+
+    # Scheduler configuration
+    scheduler: SchedulerConfig = field(default_factory=SchedulerConfig)
+
+    # Cache configuration
+    dit_cache: DiTCacheConfig = field(default_factory=DiTCacheConfig)
+
+    # Resource limits
+    max_memory_gb: float = 16.0
+    max_gpu_utilization: float = 0.8
+
+    # Timeouts
+    request_timeout: int = 300
+    worker_timeout: int = 60
+
+    def validate(self) -> None:
+        """Validate configuration parameters."""
+        if self.max_memory_gb <= 0:
+            raise ValueError("max_memory_gb must be positive")
+        if not 0 < self.max_gpu_utilization <= 1:
+            raise ValueError("max_gpu_utilization must be between 0 and 1")
+```
+
+### 5. Error Handling
+
+```python
+class CoreModuleError(Exception):
+    """Base exception for core module errors."""
+    pass
+
+class SchedulerError(CoreModuleError):
+    """Scheduler-related errors."""
+    pass
+
+class QueueFullError(SchedulerError):
+    """Queue is at capacity."""
+    pass
+
+class CacheError(CoreModuleError):
+    """Cache-related errors."""
+    pass
+
+class ResourceError(CoreModuleError):
+    """Resource allocation errors."""
+    pass
+```
+
+### 6. Examples
+
+#### Basic Usage
+```python
+from vllm_omni.core import CoreScheduler, CoreModuleConfig
+from vllm_omni.request import create_text_request
+
+# Create configuration
+config = CoreModuleConfig(
+    scheduler=SchedulerConfig(scheduler_type=SchedulerType.FIFO),
+    max_memory_gb=8.0
+)
+
+# Initialize scheduler
+scheduler = CoreScheduler(config)
+await scheduler.start()
+
+# Create and schedule a request
+request = create_text_request(
+    request_id="req_001",
+    prompt="Hello, world!",
+    sampling_params=sampling_params
+)
+
+success = await scheduler.schedule_request(request)
+if success:
+    result = await scheduler.process_request(request)
+    print(f"Result: {result}")
+```
+
+#### Advanced Usage
+```python
+# Custom scheduler with priority handling
+config = CoreModuleConfig(
+    scheduler=SchedulerConfig(
+        scheduler_type=SchedulerType.PRIORITY,
+        priority_weights={"high": 2.0, "normal": 1.0, "low": 0.5}
+    )
+)
+
+scheduler = CoreScheduler(config)
+await scheduler.start()
+
+# Monitor scheduler status
+status = scheduler.get_status()
+print(f"Queue size: {status['queue_size']}")
+print(f"Cache hit rate: {status['cache_hit_rate']:.2%}")
+```
+
+---
+
+## Module-Specific Guidelines
+
+### Core Module
+- Focus on scheduling, caching, and resource management
+- Provide clear interfaces for other modules
+- Handle concurrency and thread safety
+- Implement comprehensive monitoring
+
+### Engine Module
+- Handle model loading and inference
+- Support both AR and diffusion models
+- Provide unified interface for different model types
+- Implement efficient memory management
+
+### Executor Module
+- Coordinate between different engines
+- Handle request routing and load balancing
+- Manage execution pipelines
+- Provide error recovery mechanisms
+
+### Worker Module
+- Handle actual model execution
+- Manage GPU resources
+- Implement batching strategies
+- Provide performance optimization
+
+---
+
+## Checklist for API Design
+
+- [ ] **Clear Purpose**: Module purpose is well-defined
+- [ ] **Complete Interface**: All public methods are documented
+- [ ] **Error Handling**: Comprehensive error handling strategy
+- [ ] **Configuration**: Flexible configuration system
+- [ ] **Examples**: Basic and advanced usage examples
+- [ ] **Type Hints**: All methods have proper type hints
+- [ ] **Documentation**: Comprehensive docstrings
+- [ ] **Testing**: Unit tests for all public methods
+- [ ] **Integration**: Clear integration points with other modules
+- [ ] **Performance**: Performance considerations documented
+
+---
+
+## Submission Guidelines
+
+1. Create a new file: `docs/api/[module_name]_api.md`
+2. Follow the template structure exactly
+3. Include all required sections
+4. Provide working code examples
+5. Ensure all methods have proper docstrings
+6. Include error handling strategies
+7. Submit for review before implementation
+
+This template ensures consistency and completeness across all vLLM-omni modules.

docs/contributing/design_documents/architecture_overview.md

@@ -0,0 +1,139 @@
+# Architecture Overview
+
+# Introduction
+
+This document outlines the architectural design for vLLM-omni, a new repository developed under the vLLM project. vLLM-omni supports multi-modality models with non-autoregressive structures and non-textual outputs, extending vLLM beyond traditional text-based, autoregressive generation.
+
+# Goals
+
+The primary goals for vLLM-omni are:
+
+* **Support for Non-textual Output:** Enables the integration, efficient processing and output of various data types, including but not limited to, images, audio, and video, alongside text.  
+* **Non-Autoregressive Structure:** Support model structure beyond autoregressive, especially Diffusion Transformer (DiT), which is widely used in visual and audio generation.  
+* **Integration with vLLM Core:** Maintain compatibility and leverage existing vLLM key modules and optimizations where applicable.  
+* **Extensibility:** Design a modular and flexible architecture that can easily accommodate new modalities, model architectures, and output formats.
+
+# Expected Gains
+
+* It breaks through the limitation of vLLM that only supports autoregressive text generation models, and provides architectural support for popular model structures, especially multimodal generation.  
+* By supporting new output formats and new model architectures, developers can easily connect to mature ecosystems such as Higress Gateway and Ray, improving the ease of deployment and efficiency of corresponding models.  
+* Integrating efficient cache management, multiple parallel strategies, disaggregation from vLLM, to improve inference efficiency of corresponding models.
+
+# Non-Autoregressive Structure
+
+* **Diffusion Transformer:** initialized with a random latent noise, the transformer model iteratively predicts latent output for multiple steps, under control of sampling timestamps. Finally, it is decoded by the decoder of Variational AutoEncoder (VAE). It is not supported by vLLM now.  
+* **Autoregressive:** dominant generation paradigm for text, now supported by vLLM. It generates tokens conditioned on previous tokens one by one. The efficient KV Cache management provided by vLLM can accelerate inference effectively.  
+  ![][image1]                       ![][image2]
+
+
+|  | Autoregressive | DiT |
+| :---- | ----- | ----- |
+| Scenarios | Text generation | Multimodal generation |
+| Generation | Token-by-token, left to right | Continuous vector, iteratively refine |
+| Key advantages | Smooth, good local consistency | Global consistency, diversity and controlability |
+| Efficiency | More efficient with KV cache | Need generation repeatedly, slow |
+| Error accumulation | Easy to accumulate | Can fix possibly |
+| Key challenges | Limited global planning, long context generation | Large global computational cost |
+| Sequence length | Varied length | Fixed length |
+| Attention mask | Lower triangular | Full bi-directional matrix |
+| Parallel strategy | TP, DP, PP, .etc | DP, CP, .etc |
+| vLLM support | Supported | Unsupported |
+
+# Targeted models to support
+
+According to analysis for current popular open-source models, most of them have the combination of AR+DiT. Specifically, they can be further categorized into 3 types below:
+
+* **DiT as a main structure, with AR as text encoder (e.g.: Qwen-Image)**  
+  A popular model for image generation and editing. Main visual generation models are similar to the structure, such as Flux.  
+  ![][image3]  
+* **AR as a main structure, with DiT as multi-modal generator (e.g. BAGEL)**  
+  A unified multimodal comprehension and generation model. Visual generation can utilize CoT text generation output. Main unified multimodal models are similar to the structure.  
+      ![][image4]  
+* **Extension, multiple AR+DiT (e.g. Qwen-omni)**  
+  A new model for multimodal input and output. It designs the Thinker-Talker-Codec structure, which is a double-AR+DiT format.  
+  ![][image5]  
+
+
+
+# Architectural Overview
+
+vLLM-omni will import core modules from existing vLLM framework and introduce several new components and modifications to achieve its goals. The core architecture will be based on a modular design, allowing for independent development and integration of different modality handlers and output generators.  
+![][image6]
+
+## General ideas
+
+* It divides models in vLLM-omni into multiple stages, which are processed by different Engine Cores. Each stage can choose AR Engine Core for AR model part, and Diffusion Engine Core for diffusion model part.  
+* Designed a specialized diffusion engine core for DiT or dLLM.  
+* Plug-in with external DiT acceleration modules such as xDiT, Cache-DiT.  
+* Output linked with Higress API, Gradio for online demo, ComfyUI for less-code process.
+
+## Key Components
+
+* **Processor:** Input processing, tokenization or encoding  
+* **Scheduler:** for order and queue of incoming requests  
+* **Executor:**  Prepare model input and run  
+* **Multimodal output processor:** support multimodal outputs  
+* **Hidden State Output processor:** support intermediate hidden vectors as request output  
+* **Engine Core:** accelerate AR modules, imported from vLLM main repo  
+* **Diffusion Engine Core:** Wrap and accelerate DiT modules.  
+* **Diffusion model:** Wrap up diffusion models with vLLM format
+
+# Examples data flow of specific models
+
+1. Qwen-omni  
+   ![][image7]![][image8]  
+2. Qwen-Image  
+   ![][image9]      ![][image10]
+
+# Overview of function call dependency
+
+![][image11]
+
+# Roadmap
+
+* AsyncLLM:  
+- [ ] Initializer: resource allocation for different stages (2025 H2)  
+- [ ] Hidden state Output Processor (2025 H2)  
+- [ ] Multimodal Output Processor (2025 H2)  
+- [ ] Interface for API, Gradio, ComfyUI  
+* OmniEngine:  
+- [ ] Basic Diffusion Engine core: Diffusion step manager, Cache manager (2025 H2)  
+- [ ] Pipeline of different Engine Cores (2025 H2)  
+- [ ] Communication between Engine Cores via AsyncLLM (2025 H2)  
+* Model support (Included but not limited to):  
+- [ ] DiT: Qwen-Image (2025 H2), Flux-Kontext  
+- [ ] Unified-MLLM: BAGEL (2025 H2), OmniGen2, Janus-Pro  
+- [ ] Extensions of AR+DiT: Qwen-omni (2025 H2)  
+* Optimization  
+- [ ] Integration of other DiT acceleration modules, such as xDiT, Cache-DiT  
+- [ ] Disaggregation for AR and DiT  
+- [ ] Integration with vLLM's existing KV cache for potential future hybrid models.  
+- [ ] Support for streaming multi-modal inputs and outputs
+
+# Contact
+
+For any questions or feedback regarding this design document, please contact  
+Han Gao: [hgaoaf@connect.ust.hk](mailto:hgaoaf@connect.ust.hk),  
+Hongsheng Liu: [vllm-project@gmail.com](mailto:vllm-project@gmail.com)
+
+[image1]: <data:image/png;base64,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 truncated
+
+[image2]: <data:image/png;base64,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 truncated
+
+[image3]: <data:image/png;base64,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 truncated
+
+[image4]: <data:image/png;base64,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 truncated
+
+[image5]: <data:image/png;base64,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 truncated
+
+[image6]: <data:image/png;base64,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 truncated
+
+[image7]: <data:image/png;base64,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 truncated
+
+[image8]: <data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAVwAAAGJCAYAAADG5YuwAACAAElEQVR4XuydBXgV19aG23tv7//f+u3f21svFHd3d5fgkgRCcHeXIIFAkKDB3d1dgwa3lhYo7m4tUOT797fCnJ6chJaE5JCTs77nWU8ye/bImVnzztpr75l5CyqVSqVyit5yLFCpVCpV7EiBq1KpVE6SAlelUqmcJAWuSqVSOUkKXJVKpXKSFLgqlUrlJClwVSqVyklS4KpUKpWTpMBVqVQqJ0mBq1KpVE6SAlelUqmcJAWuSqVSOUkKXJVKpXKSFLgqlUrlJClwVSqVyklS4KpUKpWTpMBVqVQqJ0mBq1KpVE6SAlelUqmcJAWuSqVSOUkKXJVKpXKSFLgqlUrlJClwVSqVyklS4KpUKpWTpMB1Ez1//hyzZ89G+vTp8d133yFlypSoWrUqLl686Fg1zmjJkiVInDix7G/r1q3lN/yZqlWrJnXtrXTp0o7VIojHJUOGDFI/VapUmD9/Pq5du+ZYLdaUMWNGpE2bFvfv33ecpYpnUuC6iaZOnYoECRLgrbfestk//vEP5MqVCzdu3HCs/sb1448/IlGiRLZ9/fDDDzFy5EjHauH07bffhvt9tKRJkzpWC6d58+ZFOC4JEyZEvnz58PDhQ8fqsaJ//vOf+Nvf/oY7d+44zlLFMylw3UATJkywwaRGjRro0qULatasKREdy/7973/j3r17jou9MS1YsABJkiSRffP29kabNm3k5kAonTlzxrG66ObNmwIuRsT8fZb9GaQZQVvHpVKlSlK/Vq1atrKvvvoKjx8/dlwsxjVmzBgEBwc7ZVuqNysFbjzXgwcP8MknnwhA5s6dG27ekydP8P3338u8Dh06YOXKlfJ/y5Ytbc3b/v37S1n+/PltyzECZBmb+RSb5O+//76U5cmTBz///LOt+U94spwwIzj5/7vvvovy5cvj9u3btnXaq0yZMlIvderUtjJfX18p27Ztm13NP7R161aZ37x5c+zdu1fsz5rohNvXX38ty/CGZK/Vq1fLtjmvR48eUpY7d26ZXrp0qS2S/u9//4vz589j//79tmMcGBhoW8/169flRsEoe/HixfK7WYfbPXjwoK3exx9/LMfv7t27tjJV/JQCN56LMOBFzmZzZLKASKDwgv/ggw+QLl06LFq0SOYToBZc2MS+cuWKTNMOHz4s0Zk1/fe//13+/u///q9EihTXwzJGp/Z1aFYdRzHFce7cOdmWJQKf6yDMI1NQUFC47dAYuR86dMixqujEiRNS54svvnCcJdq4caPMZ36VKlCggG29b7/9tu13MNVhRd/WfOumwN9BkFrz7Otwu7///rvU+89//iP7qsCN/1LgxnMxIuMFXrhwYcdZoj179sj8ZMmSyXSOHDkEDJ6ennj06JHAwILF8ePHbemJzz77TKJERsisY0W0zIlyPoHEnKQF3G+++caWoyTQWcZtvYp69uwp9QsWLPjSjrPixYtLHUbdY8eORf369WWa0Xhk2rx5s8zPnj274yzRsWPHbMs/ffrUBlwrqqesqHbYsGEy3aRJE5m2ImYLuCzj9ij+Ft68mP5gBEwpcN1HCtx4ru3bt8sFz86xyLRz506Zb6UMrIiVAN63b5/8X7duXfnLPCNHAjCiq1y5sqQk3nvvPZkXmbHZbAGXALfk4eEhQH7ZPtmrVatWsjwjTSsifFVZnWGMlh1l/TaO2ohMR44csR0HQt4CLlMVltKkSYN33nkHFy5ckOk+ffpInUmTJsm0BVymDCwtXLhQ0hX/8z//Y+usVOC6jxS48VzMMRIKbAZbOVNGYhMnTpSOJquTaOjQoTKPEGBdRrT16tWTeadOnRJwMO3AiI+Q2rBhgwCQwGWUy0432m+//SbljH6fPXtmA26jRo1s+8T87asAl5DmsnXq1JF1vUycR4hzP7h9S1YkffLkSbvaYeLvZE6Vv5XHyF6MxFu0aCHLcr2UBdyjR4/a6hGcPLYW0CMDLo+PfdqCKRwFrvtKgRvPxeisUKFCAgJe6JcuXZJOLk5b5ti0z5Ytm20eYUBVr17dVtaxY0dbj3qKFCkEztOmTZNpNueLFi2KEiVKSEoiOsC9deuWdLJxOXbavYoyZ84s9fv27SvTjNwZiX/66acvhXXFihVlGUbCVq6XoyA4Rpkg5jxGwpQCVxUTUuC6gWbOnGmDJWHHiNG+84pDqUJDQ231vby8bPOsHKefn5+tzH60A1MMLCN4CVVrBANBx5RDdIDL/bXG4FaoUAGNGzeW5WkHDhyQOuzU4vSmTZtkmjcBC54sz5s3r0wTqi/TihUrbL+JNyUuxwclrDKORrDSGApcVUxIgesmsi50CyY09rCzM4v/sxPHGkbFIUzWOFhrmNOOHTtsy9k32wlVC240RrvdunXD1atXZT7hzHL7EQnswGMZn65yFDuoPv/883D7aW+zZs2SetZwtQEDBtiWK1myZLi6HGHxV+OLV61aJSkS++V48yhVqlS4Bx+sCNp+1IM1PMwaG9y1a1eZZh6cYqcYp5m6sMT9t5azOs2slI8++BD/pcB1IzG9wIuaTXb+tfKtgwcPRtOmTW312AT/9ddfpZ4V4XFZ5oBfFoWxnPUdx75yeZZzfZY4NpjrigyG9vsYmVn7w3QFp/nXXtZ+RAVejsclshQEfxfnE+yWuC3+Dqs+Ac06VrrFOmb2+8L9t5azRlxwvv20Kv5KgatSqVROkgJXpVKpnCQFrkqlUjlJClyVSqVykhS4KpVK5SQpcFUqlcpJUuCqVCqVk6TAValUKidJgatSqVROkgJXpVKpnCQFrkqlUjlJClyVSqVykhS4KpVK5SQpcFUqlcpJUuCqVCqVk6TAValUKidJgatSqVROkgLXiJ+DuXbtWpw061M1KpXK9aXANfrXv/6Fd/7xT/zj7/+IO/a3f+Cdv78jHyFUqVTxQwpco0/+/QmWHVmMvfdC44wd+HUPNp9bj/f+V4EbX8QPbNauXdtl7JNPPnH8CarXlAIXYcCds3MmtlzYGGds6+XNchN4/1/vO+6uykXFT7+7kpImTepYpHpNKXChwFU5RwpclQIXClyVc6TAVSlwocBVOUcKXJUCF+GBu+vmdoTe3IHQWzuw68a2CCB0lilw458UuCoFLv4AbsilTchdKBcy58iEDFkzoGTlkthxfauUb720OQyE5q/1v9jFsLKQi5sinW8/zTr2da35XL/Mt9+OAjfeSYGrUuAiPHDzl8iH/Q9244enh9BnbC8MnDEA7fu3hVeTmth2dQtqNfeGb2sfiYQJxfl75qK8ZzlMWjcB265sQZ1WPmjQoR52394p8PRtUwc1GlbH9qshmL19BsrXLIvBMwOx49pWAatn4xroMqSTLNt5cCczXRMbzqzFtstbIgXumDFjsHTp0nBlKtfQXwH38OHDSJEiBVq1aoXOnTsjffr0ePDggWO119a9e/fQtm1bx+IIUuDGvBS4CA9cRrf9Jvqj1xg/lKlWGnNDZ5toN73As7V/SwTNGSwgbtW7JTadW48iZQtj373daBvQBu0D2mLapskYNCMQ7cz0kNmDMGrJSCw9vAhjVwQjb9Hc+OHZYfQZ10tAXaVOZcwImQafFrUQbOplz5sN60+tCRfhvvcCuIsXL5Zxke+++67D3qtcRX8F3KNHj6Jdu3a26caNG+PHH3/EtGnTkDlzZlSsWFHKL168iGTJksHT0xN169bFpEmTsGnTJuzYsUNA/fjxY9SsWRNZs2bFyJEjcfPmTWTKlAnp0qXDDz/8gJ49e+LLL7/E9evXbduKTArcmJcCF+GBm6tgTizYO1ds5Y/LsOn8BhQuUwg7r2+DdxNPgeea4yslch27YjQqeHlItLrv/m54mig4VfrUSJc5nUS6y39YigIl8iN3odySRhi1eASSp0oGv5HdMXrZKOQpkhvrTq7G8AVD0SWoE0pXLhkupbDcAPfL/34poH3rrbfEeNE2atRILsYmTZqgefPmaNmyJdq0McBv3x4dO3ZEly5d0L17d/To0QO9evWCv78/AgICEBgYiEGDBiEoKAjDhw/HiBEjEBwcLFHz+PHj5cKdOnUqpk+fjpkzZ2L27NmYN28eFi5cKMBftmwZVq5ciTVr1mDt2rXYsGGDXOghISHYvn07du7cid27d2Pfvn3Yv38/Dh06JBAhNH7++WecPHkSp0+fxrlz53D+/HlcunQJV65cwY0bNwQK/Buf7fvvv3d0vXA6duwYPvroIyRMmBBfffWVnFsqQ4YMEpXyYQQe4759++LgwYO4fPkyGjZsKOeL54DHvnfv3vK3WbNmuH//PgoUKCDnjOf49u3bePLkiRzrTp06OWw9ohS4MS8FLsIDN2/RPNhsIGvlWDeeDYtit18LMVFtc4xYOBQBE/uiVa8WWHNiFfIWyYODv+1Dsx5N0K5/W8zdNdtExbPQZ0xPDJ8/FGtNnWmbJqHjwPZo2bM5TuMkvJt5Yf7uOahsItzZO2bCt3UdjFgwDCUrFRfQ2ke4H77/IUJDQ5E/f34BboIECeQdC7zYaIx2Lly4IBA7e/asAO2XX34RuJ04cUJAxwuZ0CP8jhw5IiA8cOCAGOG4d+9euUi5HV7QhOe2bduwdetWbNmyBZs3b8bGjRsFsOvWrRPg0latWiUX8/LlyyXNsWTJEixatEgAvWDBAsyfPx9z584VcBPgM2bMkGhtypQpmDx5sgB+4sSJmDBhgoB+3Lhx8dq+++47R9cLJ54f3jQpRqKEKcXolseIx5jnkhEqzysBypsvgctztn79egEuz2OdOnXkJsrzRvF8MXrm+eWNztrOn0mBG/NS4MIOuCYKbd2nZTjgbjq3QXK4jDy3XwlBtXpVJe/KEQzMu07eMAnFKxTDhDXjJO/q2agGKvlUlBzuZrNs1bpVJP/LacKadRnpMirm/MqmbruAtrKujoHtbR1okXWaEYDM8RGYKtLine truncated
+
+[image9]: <data:image/png;base64,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 truncated
+
+[image10]: <data:image/png;base64,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 truncated
+
+[image11]: <data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAABM8AAARlCAYAAACukOBGAACAAElEQVR4Xuy9B7hV13WuHSd5ch3Zia17rcQl13+K88dyXOLYlmPLRb13CVUkUEMIEB3Rkeiig+iiI3rvvffeO4gmeu+dM+56xz4TbRZ7IxDt7M33Ps98zll97rXGnGvOb4055l+YEEIIIYQQQgghhBAiJX8RXyGEEEIIIYQQQgghhEgg8UwIIYQQQgghhBBCiDRIPBNCCCGEEEIIIYQQIg0Sz4QQQgghhBBCCCGESIPEMyGEEEIIIYQQQggh0iDxTAghhBBCCCGEEEKINEg8E0IIIYQQQgghhBAiDRLPhBBCCCGEEEIIIYRIg8QzIYQQQgghhBBCCCHSIPFMCCGEEEIIIYQQQog0SDwTQgghhBBCCCGEECINEs+EuAhnz5jlKCnF09k0Kb49/B9fH98/1bazSfaXYhvr025PPm98e3xd8nJ826VcK36eVNeN/e/nih+Xajl+vaT1550jvu+lbIv/TfV/fDn5nPHz5yYhhBBCCCFE9iHxTIh05JgdP5Bjx3YrKcXSntTpeG5KXk61/lzafZFtsXPE07n1KfJz7pwp8nvB9ZL2uWBb8j7ptuWm+DWT973geqmOS943zfW+7Den+j++HD9Hym25/8eXk895wfmj8506mhOvRYQQQgghhBBZgMQzIdKBeLY/x45sz7GjO5SUvkhH0qT49vB/fH18/1TbLnbclx2bvC2+Pb4ueTm+Lb5Pqm2p9onvG///Upfj10ten25b/P/4cvzcqbalukb8nBecf7vEMyGEEEIIIbIViWdCXIQT+y3ROd6ppKSklD5RT0g8E0IIIYQQIjuReCZEOnIQz3ISHigpOstKSkpKIUk8E0IIIYQQInuReCZEOsKwTYlnSkpKX5IkngkhhBBCCJG9SDwTIh3yPFNSUrrEJPFMCCGEEEKI7EXimRDpkOeZkpLSJSaJZ0IIIYQQQmQvEs+ESIfEMyUlpUtMEs+EEEIIIYTIXiSeCZEOiWdKSkqXmCSeCSGEEEIIkb1IPBPiIpzYbxLPbmA6vjt6CEej57Dni+dgh83OHkgsu2CxL9rnpPlflk9wzIlon4Nf7HNyb2IfO56bonMe22UXXC+euD7nuZR9SX79KK/k8cuOYd/TkX3lHJKNZUOSeCaEEEIIIUT2IvFMiHRowoAbmhC8tq3bby0afWKfLd3mQhPrenToZ2MGT3ax7PRBs3lTllmJImVsxbz1dvpQ4pjypSrb8H7jfDv7LZu91t59s5gVzk1Vy9ew3RtQuOycoIZQhuCFsBYEtu1r91vfbkNs/+fsYC7K2ZGEKOYiGeuOJY4NQt6GKK9d2va0PZuOfnGu4wkB0MXAIOBFl58zabGNHDDBBTpSOD/58GXyd+ILYZB74PvkCm5nDnyxHL9/Stc3STwTQgghhBAie5F4JkQ6NGzzhiaEorWLt9hD9z9mcycvdVGJde++Ucwa1m7m4hMi1IAew+373/uBNW/Y1h/bkN6jfblRneYuPuHZNbDHCLvvrgds5viFtmDaUlsya5Xt2XjE5kxaZD079bcB3YfZwa2nXOBCdOvZaUC03zL7fNVuGzFgvK1bssUmjphhndv0sCmjZrvnG/v36NjPBvYc4fk9sSfHBa/5U5ZZwfxv2cblO23RjOXWu8tA69NlkO3//IQd3n7Gl7u172N7N+2zSmWq2aMPPm57Nh2wxTNXW6fW3WzCsOkuni2ZtSbK93Dr3qFvlI89djr6vRuWbY/y0N1mTljowtyaRZs9T3OnLPU8xe+h0vVLEs+EEEIIIYTIXiSeCXERNGzzxiWEMjzO8j39ki2YtsLFMtaVKVbeWjRs66IYAtKwvmPtyUeftVJFy9mhHcetecM2lu+pF619y09dcEPQGtpnjL30XH4XoXatP2CHo/3WR+f+nzvudE+0P995t/XpOtjGD5saXe9Fe+O1QvbgfY/YzAkLrGzxCtamWQe741f/YyXeLWP3/Pk+Wz5njbVq3N5KFStnBV9509o07eBeYORp0fSV5zzh7r/nQXvztXeiY+73/HRq3d3ee6ekVS73gQtolcpWs8ceftKFOfLfpF5Le+bxfDZ93Hx75YUC9nz025945Gmr8n4N2711X/S3ur1fspIVi84xacxUK/1eeav9QT17KV9+mzNxkZ2WgHbDksQzIYQQQgghsheJZ0JcBIlnNy5dqng2uNcoK1mkjL1fqrIN6D7CalerbzUq17HWTTu4eMZ+YwZNstv/4z8t/wsFrMDLr1vLxm1s7aLP7bmnXrAd6w5Yp1bdrH7NJlauRAUfErp1zT4r9EYRmzZmrpUrWTHa/xN7r3ApP1fZaJ9P2/exJx992lo1aW8fVKhlTz/+nB3YctLzE8SzZXPWuaj12dKt1rfrYKsV5avtx52sepS3kQNG277NB21g9+HWqM7HdmLP2SjvQ21Qr5H2wrMvW99Ph1iRQiVsanT9GePmW8WyVaPtw/28eNMd2XPUOrfuYX+68y4b2meYPfLA4+5pR/7i91Hp+iSJZ0IIIYQQQmQvEs+ESIeGbd7QhFC2bulWe+HZV1zock4kxLMOLbuee0yDeo6wmlU/cq8uvLVqVatnrZu0TwzjzPU8Q2B7u0DhRKyxXH1j3eIt7nW2c/2h6HyfutcXItzEETNt+7oDVrRQiYR4VqKifdK8k4tfDKdkqGW39n3tuadfsJpVPrKPG7T0oZOHtp2xnMNfiGcMu+TvphU7rVfHAVbnw4Z2cNux6Nge7s3W9ZNePuyzWb1WNnviYnv5+VetWf1W9uC9D/tQ0MrlPrSF01fYhOHTo99X14eSFn+39Lk4bd2jPNz9p/tcCKxfs7FNGjnTTu298D4qXZ8k8UwIIYQQQojsReKZEOmQeHZDE+LZ+mXb7Q+/v8uFq95detnMCXOtyFvF7eV8r1qvzr1sxMARLnzh/TV2yBS77Tu32bC+Y6xFo7ZWv1ZT9wTDGwvx7Df//VsX2Lq1722Deg6xuZOXWIGX37Bta/db8wZtrHHdFja490h7q0Bhq/NBA3vmiXw2bew8K/p2cR8KWrFMNc8XotqwfmPdY6xhnWY+bJLjT+xNzARKzLO3Cha2hdNX+vDPz5ZstS5te9iHlWpbuxado3O19mM/ad7Z+n06xAq88oa1b9HFr9e/21B75IHHXIwrWbSsTygwatBEK1uivK2cv87PhyhYqlhZ69Wlr4tzvToPsBLvlrbJo2b5PYvfR6XrkySeCSGEEEIIkb1IPBMiHRLPbmg6vjvH9m85YZ+262N1PmxgtarWs8G9R3nA/no1Gvsyccfw2po1YaHt3njExadd6w/ZwmkrfBbOk/uYtTPHA+03rdfSaler555pzeq3ttULNrlXF9fAW2zBtOW2d9Mxa1i7oRV45U2PZcZxU0fPskUzVvrwSfJFfLItq/b49Zi4oEGtprZpxS47tS8xacC2Nfv8vMzUOX7YNNuz8agtn7POxbrN0X51qzf0vByIrrt55S6Pp8YQT2boRFQbPWiSX2/K6Nm2dfVej82GMMaEALMmLvJhqUP6jLacI4nZOvk9vToNiOz0rN+z+H1Uuj5J4pkQQgghhBDZi8QzIdIh8eyGJ4ZJMlTTTuemI5aYZTMsn0gMy8yJ0ond5sMZmTGT4P2ITTw70sm9SceQTpqd2pc4F9dgf87F7JYli5axd94o4gIY52afMBkAeWKZ83EdO5VILIdr+frDudujv8d3J2K1cS7PB8ecTKz35ZOJ7f47o//5jZ6fQ4lznGKfQ4nze5y33PtwlGXyx/KxxO+I3z+l65ckngkhhBBCCJG9SDwTIh0Sz266hACFCIewpiGQSpeTJJ4JIYQQQgiRvUg8EyIdEs9uyoSAJi8upctNEs+EEEIIIYTIXiSeCZEOiWdKSkqXmCSeCSGEEEIIkb1IPBMiHRLPlJSULjFJPBNCCCGEECJ7kXgmRDoknikpKV1ikngmhBBCCCFE9iLxTIh0SDxTUlK6xCTxTAghhBBCiOxF4pkQ6ZB4pqSkdIlJ4pkQQgghhBDZi8QzIdIh8UxJKU8kZj89nsdnQJV4JoQQQgghRPYi8UyIdCCe7Ys6xdsTHWMlJaXrn45G6cRus5N77YJteSpF9cSpIxLPhBBCCCGEyEYkngmRhpyoH4wnCR1iJaXz0lG7cDms+7JtydtTLce3pTrvxY5LtRzfFj82+RrxY+P7xrelOzbVvun+j58nadvZE2Zd2/ewMsUr2IlDZ+x0/Nj4NeLnSZe/VMelOybVtvj/UTpzQuKZEEIIIYQQ2YjEMyGEEHmacePHWdu2bS0n52x8kxBCCCGEEEJccySeCSGEEEIIIYQQQgiRBolnQgghhBBCCCGEEEKkQeKZEEIIIYQQQgghhBBpkHgmhBAiTzNixAj76KOP7OxZxTwTQgghhBBCXH8kngkhhMjTFC9e3G699VY7ffp0fJMQQgghhBBCXHMkngkhhMjTVKxY0f7t3/7NTp06Fd8khBBCCCGEENcciWdCCCHyNJUqVbJ//dd/lXgmhBBCCCGEuCFIPBNCCJGnqVy5sv3Lv/yLxDMhhBBCCCHEDUHimRBCiDwNwzZ/+MMfSjwTQgghhBBC3BAkngkhhMjT4Hkm8UwIIYQQQghxo5B4JoQQIk+za9cuW7NmjeXk5MQ3CSGEEEIIIcQ1R+KZEEIIIYQQQgghhBBpkHgmhBBCCCGEEEIIIUQaJJ4JIYQQQgghhBBCCJEGiWdCCCHyNK1bt7Z77rlHEwYIIYQQQgghbggSz4QQQuRpqlSpYn//939vJ0+ejG8SQgghhBBCiGuOxDMhhBB5mpo1a9p3v/tdiWdCCCGEEEKIG4LEMyGEEHmaWrVq2T/+4z9KPBNCCCGEEELcECSeCSGEyNMgnv3DP/yDxDMhhBBCCCHEDUHimRBCiDxNtWrV7JZbbpF4JoQQQgghhLghSLine truncated

docs/contributing/design_documents/index.md

@@ -0,0 +1,12 @@
+# Design Documents
+
+This section contains design documents and architecture specifications for vLLM-omni.
+
+## Architecture Documents
+
+- [vLLM-omni Software Design Document](vllm_omni_design.md)
+- [Architecture Overview](architecture_overview.md)
+
+## API Design Documentation
+
+- [vLLM-omni API Documentation](api_design_doc.md)

docs/contributing/design_documents/vllm_omni_design.md

@@ -0,0 +1,509 @@
+# Architecture Design Principles
+
+## Overview
+
+vLLM-omni is a multi-modality extension for vLLM that supports non-autoregressive structures and non-textual outputs. This document outlines the key software abstractions, APIs, and dependencies for the system, designed to maximize reuse of vLLM's proven architecture.
+
+## Architecture Principles
+
+1. **vLLM V1 Compatibility**: Built on vLLM's Engine V1 architecture with AsyncLLM and EngineCore patterns
+2. **Stage-based Processing**: Models are divided into multiple stages, each processed by different Engine Cores
+3. **Multiple Engine Core Support**: Each stage can use either AR Engine Core (reusing vLLM) or Diffusion Engine Core (new DiT support), or other new Cores
+4. **Worker Process Pattern**: Follows vLLM's multiprocess worker architecture for scalability
+5. **Extensibility**: Easy integration of new modalities, model architectures, and output formats
+
+## Key Data Flow
+
+API Server --> OmniLLM/AsyncOmniLLM (New, including multi engines) --> LLMEngine/AsyncLLM --> Engine Core
+ --> Scheduler (New one for DiT) --> Executor (New one for diffusers) --> Worker (New one for DiT)
+ --> ModelRunner (New one for AR hiddenstate, New one for DiT) --> RequestState --> OutputProcessoer (New one for final multimodal output)
+
+## Core Components (Ordered by Data Flow)
+
+### 1. Installation
+```bash
+pip install vllm
+pip install vllm-omni
+```
+
+### 2. Online inference launch (Entry Point)
+
+Keep consistency with the vllm main branch
+```bash
+vllm serve Qwen/Qwen2.5-Omni-7B --omni --port 8000
+```
+
+#### Design Architecture
+```mermaid
+graph TD
+    A[vllm serve model --omni] --> B[vLLM-omni CLI Wrapper]
+    B --> C{Detect --omni flag}
+    C -->|Yes| D[Parse OmniConfig]
+    C -->|No| E[Forward to vLLM CLI]
+    D --> F[Initialize AsyncOmniLLM]
+    F --> G[Start omni Server]
+    G --> H[Multi-stage Processing]
+    E --> I[Standard vLLM Pipeline]
+
+    subgraph "vLLM-omni Components"
+        F
+        G
+        H
+    end
+
+    subgraph "vLLM Components"
+        I
+    end
+```
+
+#### omni Serve Command Implementation
+File: vllm_omni/entrypoints/cli/main.py
+```python
+import sys
+import argparse
+from typing import List, Optional
+from vllm_omni.entrypoints.omni import OmniServeCommand
+
+def main():
+    """Main CLI entry point that intercepts vLLM commands"""
+    # Check if --omni flag is present
+    if "--omni" in sys.argv:
+        # Remove --omni flag and process with vLLM-omni
+        omni_args = [arg for arg in sys.argv[1:] if arg != "--omni"]
+        omni_serve = OmniServeCommand()
+        omni_serve.run(omni_args)
+    else:
+        # Forward to original vLLM CLI
+        from vllm.entrypoints.cli.main import main as vllm_main
+        vllm_main()
+
+if __name__ == "__main__":
+    main()
+```
+
+#### vLLM Plugin Integration
+File: vllm_omni/pyproject.toml
+```python
+[project.scripts]
+# Override vLLM's CLI entry point
+vllm = "vllm_omni.cli:main"
+
+# Add entry point for vLLM integration
+[project.entry-points."vllm.plugins"]
+omni = "vllm_omni.plugin:OmniPlugin"
+```
+
+### 3. Offline inference launch (Entry Point)
+Design an upper level class to incorporate multi Engines, each engine has a engine call.
+```python
+from vllm.entrypoints.llm import LLM
+from vllm.v1.engine.llm_engine import LLMEngine
+
+class OmniLLM(LLM):
+    """Extended LLM supporting multiple engines and stage-based processing"""
+
+    def __init__(self, stage_configs: List[StageConfig]):
+        super().__init__()
+        self.stage_configs = stage_configs
+        self.engine_list = []  # List of AsyncLLM instances for each stage
+        self.output_processor = MultimodalOutputProcessor()
+        self._initialize_stage_engines()
+
+    def _initialize_stage_engines(self) -> None:
+        """Initialize LLMEngine instances for each stage"""
+        for stage_config in self.stage_configs:
+            stage_llm = LLMEngine.from_vllm_config(
+                    vllm_config=stage_config.vllm_config,
+                    executor_class=self.executor_class,
+                    log_stats=self.log_stats
+                )
+            self.engine_list.append(stage_llm)
+
+    def generate(
+        self,
+        stage_args_list: List[stage_args],
+        use_tqdm: Union[bool, Callable[..., tqdm]] = True,
+        lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
+        priority: Optional[list[int]] = None,
+    ) -> list[RequestOutput]:
+        """Main generation interface - orchestrates multi-stage processing"""
+
+        # Process through each stage sequentially
+        for i, stage_config in enumerate(self.stage_configs):
+            stage_engine = self.engine_list[i]
+
+            # Prepare input for this stage
+            stage_args = stage_args_list[i]
+            prompt_str, engine_request, tokenization_kwargs = self._process_stage_inputs(stage_config, **stage_args)
+
+            # Add inputs to Engine
+            stage_engine.add_request(requesy_id, prompt_str, tokenization_kwargs)
+            # Run Engine
+            stage_output = stage_engine.step()
+
+            # Update stage input and output of each stage for later usage
+            self._update_stage_io(stage_output, stage_config)
+
+
+        # Process final output
+        final_output = self.output_processor.process_output(
+            self.stage_configs[-1].stage_output
+        )
+        return final_output
+
+    def _process_stage_inputs(stage_config, **stage_args) -> Any:
+        """Prepare input for specific stage"""
+        if stage_config.engine_type == "AR":
+            return self._process_ar_inputs(**stage_args)
+        elif stage_config.engine_type == "DiT":
+            return self._process_dit_inputs(**stage_args)
+        else
+            raise NotImplementedError
+
+    def _process_dit_inputs(**stage_args)-> Any:
+        image_latent = self.vae.encode(**stage_args)
+
+        return image_latent
+```
+
+### 4. Stage config for setting of different model stages
+```python
+@dataclass
+class StageConfig:
+    """Configuration for a processing stage"""
+    stage_id: int
+    engine_type: str  # "AR" or "DiT"
+    model_path: str
+    input_modalities: List[str]
+    output_modalities: List[str]
+    vllm_config: Optional[VllmConfig] = None  # For engine config of corresponding stage
+    executor_class: type[Executor]  # For execute class config of corresponding stage
+    dit_config: Optional[DiTConfig] = None  # For diffusion stages
+    cache_config: Optional[DiTCacheConfig] = None
+```
+For AR stage, the setting is:
+```python
+from vllm.v1.core.sched.scheduler import Scheduler
+from vllm.v1.executor.multiproc_executor import MultiprocExecutor
+ar_stage_config = StageConfig()
+ar_stage_config.vllm_config.scheduler_config.scheduler_cls = Scheduler # original vllm scheduler for AR
+ar_stage_config.executor_class = MultiprocExecutor # original vllm executor for AR
+```
+For DiT stage, the setting is:
+```python
+from vllm.v1.executor.multiproc_executor import MultiprocExecutor
+dit_stage_config = StageConfig()
+dit_stage_config.vllm_config.scheduler_config.scheduler_cls = DiffusionScheduler # New scheduler for DiT
+
+# For diffusion models without using diffusers
+dit_stage_config.executor_class = MultiprocExecutor
+
+# For diffusion models using diffusers
+dit_stage_config.executor_class = DiffusersPipelineExecutor
+```
+
+### 5. Online inference main class
+
+Similar to OmniLLM in offline inference, add some asynchronous processing, referring to AsyncLLM
+```python
+from vllm.v1.engine.async_llm import AsyncLLM
+
+
+class AsyncOmniLLM(AsyncLLM):
+    """Extended AsyncLLM supporting multiple engines and stage-based processing"""
+
+    def __init__(self, stage_configs: List[StageConfig]):
+        super().__init__()
+```
+
+### 6. Engine Core
+No need to change. The specific executor, scheduler and output type can be transferred into it with new configs.
+```python
+# class: EngineCore
+# EngineCore.step (simplified)
+class EngineCore:
+    def step(self):
+        scheduler_output = scheduler.schedule()
+        model_output = executor.execute_model(scheduler_output)
+        engine_outputs = scheduler.update_from_output(
+            scheduler_output, model_output
+        )
+        return engine_outputs
+```
+
+### 7. Scheduler
+Create a new Diffusion Scheduler for DiT, which is inherited from original vllm AR scheduler. At first, no complicated strategy.
+Just process the request first in first out. Then create a class of DiT Cache Manager for optimization in the future.
+```python
+from vllm.v1.core.sched.scheduler import Scheduler
+class DiffusionScheduler(Scheduler):
+
+    def __init__(
+        self,
+        vllm_config: VllmConfig,
+        kv_cache_config: KVCacheConfig,
+        dit_cache_config: DiTCacheConfig,
+        structured_output_manager: StructuredOutputManager,
+        mm_registry: MultiModalRegistry = MULTIMODAL_REGISTRY,
+        include_finished_set: bool = False,
+        log_stats: bool = False,
+    ) -> None:
+
+    self.dit_cache_manager = DiTCacheManager(dit_cache_config)
+```
+For DiT Cache Manager (Can refer to xDiT):
+```python
+class DiTCacheManager:
+    """Manages DiT-specific caching"""
+
+    def __init__(self, config: DiTCacheConfig):
+        self.cache_tensors: Dict[str, torch.Tensor] = {}
+        self.cache_groups: List[DiTCacheTensor] = config.dit_cache_tensors
+
+    def allocate_cache(self, request_id: str, size: int) -> torch.Tensor
+    def get_cache(self, request_id: str) -> Optional[torch.Tensor]
+    def release_cache(self, request_id: str) -> None
+    def clear_expired_cache(self) -> None
+```
+
+### 8. Executor for DiT without diffusers
+
+No need to change. Just import original AR executor.
+
+### 9. Diffusers Pipeline Executor (no worker)
+
+A single-process executor that directly runs the Diffusers pipeline without spawning workers or using RPC. Interfaces remain identical to `Executor` so the EngineCore loop is unchanged.
+
+#### Function map（Pipeline Executor）
+
+##### 1) Inherited and overridden
+```python
+from concurrent.futures import Future
+from typing import Optional, Union, Callable, Any
+import torch
+import torch.nn as nn
+from vllm.v1.executor.abstract import Executor
+from vllm.v1.kv_cache_interface import KVCacheConfig, KVCacheSpec
+from vllm.v1.outputs import ModelRunnerOutput
+from vllm.tasks import SupportedTask
+from diffusers import DiffusionPipeline
+
+class DiffusersPipelineExecutor(Executor):
+    supports_pp: bool = False  # Single-process, no TP/PP/DP
+
+    def _init_executor(self) -> None:
+        # Called by ExecutorBase.__init__
+        self._failure_callback: Optional[Callable[[], None]] = None
+        self._device = self._resolve_device()
+        self._dtype = self._resolve_dtype()
+        self._pipeline = self._build_pipeline(device=self._device, dtype=self._dtype)
+        self._profiler = None
+        self._is_failed = False
+        self.is_sleeping = False
+        self.sleeping_tags: set[str] = set()
+
+    # major functions to build/run diffusers pipeline
+    def _build_pipeline(self, device:torch.device, dtype:torch.dtype)->DiffusionPipeline:
+        model_name = "Qwen/Qwen-Image"
+
+        self.pipe = DiffusionPipeline.from_pretrained(model_name, torch_dtype=dtype)
+        self.pipe = pipe.to(device)
+
+    def _run_pipeline(self, scheduler_output) -> ModelRunnerOutput: ...
+        positive_magic = {
+            "en": ", Ultra HD, 4K, cinematic composition.", # for english prompt
+            "zh": ", 超清，4K，电影级构图." # for chinese prompt
+        }
+
+        # Generate image
+        prompt_embeds = self._get_and_process_prompt_embeds(scheduler_output, positive_magic)
+        negative_prompt_embeds = self.pipe.embed_prompt(" ")
+
+        width, height = aspect_ratios["16:9"]
+
+        image = pipe(
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
+            width=width,
+            height=height,
+            num_inference_steps=50,
+            true_cfg_scale=4.0,
+            generator=torch.Generator(device="cuda").manual_seed(42)
+        ).images[0]
+
+        output = self.wrap_image_as_ModelRunnerOutput(image)
+        return output
+
+    # ---- Internal helpers (implementation-specific, not public API) ----
+    def _resolve_device(self): ...
+    def _resolve_dtype(self): ...
+    def _get_model(self) -> nn.Module: ...
+    def _get_and_process_prompt_embeds(self, scheduler_output, positive_magic):
+        ...
+        #append the positive_magic to prompt and embed them to prompt embed tensors
+
+    # Functions related to workers should either raise NotImplementedError
+    # or return empty defaults in this no-worker executor.
+
+    def collective_rpc(self, method, timeout=None, args=(), kwargs=None) -> list[Any]:
+        # No workers in pipeline executor
+        raise NotImplementedError("No workers in DiffusersPipelineExecutor")
+
+    def initialize_from_config(self, kv_cache_configs: list[KVCacheConfig]) -> None:
+        return  # no-op (pipeline already built in _init_executor)
+
+    def register_failure_callback(self, callback):
+        self._failure_callback = callback
+
+    def determine_available_memory(self) -> list[int]:  # bytes
+        # Single device; return [available_bytes]. If CPU-only, return [0].
+        return [self._determine_available_bytes(self._device)]
+
+```
+
+### 10. Worker
+
+#### Inheritance strategy
+Prefer reusing the mature GPU Worker end-to-end. Worker class is selected by configuration (vllm_config). Do not add a new executor-specific worker binding. If customization is needed, override only `init_device` to construct the `DiffusionModelRunner`; all other behaviors (device init details, profiling, sleep/wake, PP/TP comms, execute path) remain from `vllm/v1/worker/gpu_worker.py::Worker`.
+
+
+Inherited and overridden
+```python
+# Optional: only if you need to plug a custom DiffusionModelRunner.
+from vllm.v1.worker.gpu_worker import Worker as GPUWorker
+from vllm.v1.worker.diffusion_model_runner import DiffusionModelRunner
+
+class Worker(GPUWorker):
+    def init_device(self) -> None:
+        #those related to device check and init
+        ...
+
+        self.model_runner = DiffusionModelRunner(self.vllm_config, ...)
+```
+
+### 11. Model Runner
+
+### Function map (Model Runner)
+#### 1) Inherited and overridden
+Those parts relied to the KV Cache will be omitted if we do not register the model to the vllm config. The engine core will view it as do not require KV Cache, and handle it properly
+
+Reuse `vllm/v1/outputs.py::ModelRunnerOutput`：
+- DiffusionModelRunner: Use the `pooler_output=[Tensor,...]` to return multi modal tensors
+- ARModelRunner: Use the `pooler_output=[Tensor,...]` to return hidden states.
+```python
+from typing import Optional, Union
+import torch
+from vllm.v1.worker.gpu_model_runner import GPUModelRunner
+from vllm.v1.outputs import ModelRunnerOutput
+
+
+class DiffusionModelRunner(GPUModelRunner):
+    @torch.inference_mode()
+    def execute_model(
+        self,
+        scheduler_output: "SchedulerOutput",
+        intermediate_tensors: Optional[IntermediateTensors] = None,
+    ) -> Union[ModelRunnerOutput, IntermediateTensors]:
+        ...
+        return ModelRunnerOutput(
+            req_ids=[...],
+            req_id_to_index={...},
+            sampled_token_ids=[],
+            spec_token_ids=None,
+            logprobs=None,
+            prompt_logprobs_dict={},
+            pooler_output=[Tensor,...],         # Return Hidden states
+            kv_connector_output=None,
+            num_nans_in_logits=None,
+        )# return multi modal tensors via pooler_output=[Tensor,...]
+
+
+class ARModelRunner(GPUModelRunner):
+    @torch.inference_mode()
+    def execute_model(
+        self,
+        scheduler_output: "SchedulerOutput",
+        intermediate_tensors: Optional[IntermediateTensors] = None,
+    ) -> Union[ModelRunnerOutput, IntermediateTensors]:
+        ...
+        return ModelRunnerOutput(
+            req_ids=[...],
+            req_id_to_index={...},
+            sampled_token_ids=[],
+            spec_token_ids=None,
+            logprobs=None,
+            prompt_logprobs_dict={},
+            pooler_output=[Tensor,...],         # Return Hidden states
+            kv_connector_output=None,
+            num_nans_in_logits=None,
+        )# return hidden states via pooler_output=[Tensor,...]
+```
+
+### 12. RequestState Role and Usage
+```bash
+# RequestState serves as the per-request state tracker in OutputProcessor:
+# - Maintains request-specific state (tokens, logprobs, detokenizer, etc.)
+# - Converts EngineCoreOutput → RequestOutput/PoolingRequestOutput
+# - Manages request lifecycle from registration to completion
+AsyncLLM.add_request()
+↓
+OutputProcessor.add_request()
+↓
+RequestState.from_new_request() → 创建请求状态
+
+AsyncLLM.__init__() / AsyncLLM.generate() / AsyncLLM.encode() →  创建一个Background loop 持续从EngineCore获取输出
+↓
+OutputProcessor.process_outputs() → 更新状态并处理输出
+↓
+RequestState.make_request_output() → 转换为最终输出,格式为RequestOutput或者PoolingRequestOutput
+↓
+RequestOutputCollector.put() → 推送到队列（AsyncLLM）
+```
+
+Need to add implementation for an existing method
+```python
+class RequestState:
+    def __init__(self, request_id: str, parent_req: Optional[ParentRequest],
+                 request_index: int, lora_name: Optional[str],
+                 output_kind: RequestOutputKind, prompt: Optional[str],
+                 prompt_token_ids: list[int], logprobs_processor: Optional[LogprobsProcessor],
+                 detokenizer: Optional[IncrementalDetokenizer],
+                 max_tokens_param: Optional[int], arrival_time: float,
+                 queue: Optional[RequestOutputCollector], log_stats: bool):
+
+    def _new_pooling_output(self, pooling_output: torch.Tensor) -> PoolingOutput:
+        """Create PoolingOutput for multimodal/pooling requests"""
+        return PoolingOutput(data=pooling_output)
+```
+
+
+### 13. OutputProcessor
+
+For hidden state output, original OutputProcessor already support.
+
+We only need to add new one for final multimodal output.
+```python
+class MultimodalOutputProcessor(OutputProcessor):
+    """Handles multimodal output processing"""
+
+    def __init__(self):
+        self.output_handlers: Dict[str, OutputProcessor] = {}
+
+    def process_outputs(self, engine_core_outputs: list[EngineCoreOutput], ...):
+        for engine_core_output in engine_core_outputs:
+            # Option 1: Use output_type field (if EngineCoreOutput is extended)
+            if engine_core_output.output_type == "image":
+                self._process_image_output(engine_core_output)
+            elif engine_core_output.output_type == "text+image":
+                self._process_text_image_output(engine_core_output)
+            elif engine_core_output.output_type == "latents":
+                self._process_latents_output(engine_core_output)
+            elif engine_core_output.output_type == "text":
+                self._process_text_output(engine_core_output)
+            else:
+                # Fallback: use existing pooling_output logic
+                if engine_core_output.pooling_output is not None:
+                    self._process_pooling_output(engine_core_output)
+                else:
+                    self._process_text_output(engine_core_output)
+```

docs/contributing/model/README.md

@@ -1,2 +0,0 @@
-# Adding a New Model
-Stay tuned for guidance on how to add a new model on vLLM-Omni!

docs/design/architecture_overview.md

@@ -1,195 +0,0 @@
-# Architecture Overview
-
-This document outlines the architectural design for vLLM-Omni.
-
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/omni-modality-model-architecture.png">
-    <img alt="Omni-Modality Model Architecture" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/omni-modality-model-architecture.png" width=55%>
-  </picture>
-</p>
-
-# Goals
-
-The primary goal of the vLLM-Omni project is to build the fastest and easiest-to-use open-source Omni-Modality model inference & serving engine. vLLM-Omni extends the original vLLM, which was created to support large language models for text-based autoregressive (AR) generation tasks. vLLM-Omni is designed to support:
-
-* **Non-textual Output:** Enables the integration, efficient processing and output of various data types, including but not limited to, images, audio, and video, alongside text.
-* **Non-Autoregressive Structure:** Support model structure beyond autoregressive, especially Diffusion Transformer (DiT), which is widely used in visual and audio generation.
-* **Integration with vLLM Core:** Maintain compatibility and leverage existing vLLM key modules and optimizations where applicable.
-* **Extensibility:** Design a modular and flexible architecture that can easily accommodate new modalities, model architectures, and output formats.
-
-
-# Representative omni-modality models
-
-According to analysis for current popular open-source models, most of them have the combination of AR+DiT. Specifically, they can be further categorized into 3 types below:
-
-**DiT as a main structure, with AR as text encoder (e.g.: Qwen-Image)**
-  A powerful image generation foundation model capable of complex text rendering and precise image editing.
-
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/ar-main-architecture.png">
-    <img alt="Qwen-Image" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/ar-main-architecture.png" width=30%>
-  </picture>
-</p>
-
-**AR as a main structure, with DiT as multi-modal generator (e.g. BAGEL)**
-  A unified multimodal comprehension and generation model, with cot text output and visual generation.
-
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/dit-main-architecture.png">
-    <img alt="Bagel" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/dit-main-architecture.png" width=30%>
-  </picture>
-</p>
-
-**AR+DiT (e.g. Qwen-Omni)**
-  A natively end-to-end omni-modal LLM for multimodal inputs (text/image/audio/video...) and outputs (text/audio...).
-
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/ar-dit-main-architecture.png">
-    <img alt="Qwen-Omni" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/ar-dit-main-architecture.png" width=30%>
-  </picture>
-</p>
-
-# vLLM-Omni main architecture
-
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/vllm-omni-main-architecture.png">
-    <img alt="vLLM-Omni Main Architecture" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/vllm-omni-main-architecture.png" width=55%>
-  </picture>
-</p>
-
-## Key Components
-
-| Component | Description |
-|-----------|-------------|
-| **OmniRouter** | provide an intelligent router for Omni-modality requests dispatch |
-| **EntryPoints** | define the APIs for offline/online serving (APIServer, Omni/AsyncOmni) and provide the OmniStage abstraction for different AR/DiT stages |
-| **AR** | adapted for omni-modality models while inheriting efficient features from vLLM, such as cache management |
-| **Diffusion** | natively implemented and optimized using acceleration components |
-| **OmniConnector** | supports fully disaggregation based on E/P/D/G (Encoding/Processing/Decoding/Generation) disaggregation across stages |
-
-Disaggregated stages are managed through configuration, such as in the Qwen3-Omni example, where stages like Thinker, Talker, and Code2wav are defined as separate OmniStage instances with specific resources and input/output type.
-
-## Main features
-
-vLLM-Omni aims to be fast, flexible, and easy to use with the following features:
-
-### Performance and Acceleration
-
-The framework achieves high performance through several optimization techniques:
-
-* **Efficient AR Support:** Leverages efficient KV cache management inherited from vLLM.
-* **Pipelined Execution:** Uses pipelined stage execution overlapping to ensure high throughput.
-* **Full Disaggregation:** Relies on the OmniConnector and dynamic resource allocation across stages.
-* **Diffusion Acceleration:** Includes integrated support for diffusion acceleration. This is managed by the acceleration layer, which handles:
-    * **Cache:** Includes DBCache, TeaCache and third-party integration(e.g., [cache-dit](https://github.com/vipshop/cache-dit)).
-    * **Parallelism:** Supports TP, CP, USP, and CFG.
-    * **Attention:** Provides an interface for third-party integration (e.g., FA3, SAGE, MindIE-SD).
-    * **Quantization:** Supports various quantization implementations including FP8 and AWQ.
-    * **FusedOps:** Allows for custom and third-party integration.
-
-### Flexibility and Usability
-
-vLLM-Omni is designed to be flexible and straightforward for users:
-
-* **Heterogeneous Pipeline Abstraction:** Manages complex model workflows effectively.
-* **Hugging Face Integration:** Offers seamless integration with popular Hugging Face models.
-* **Distributed Inference:** Supports tensor, pipeline, data, and expert parallelism.
-* **Streaming Outputs:** Supports streaming outputs.
-* **Unified API:** Provides a consistent and unified API interface compatible with vLLM.
-* **OpenAI-compatible API Server:** Includes a FastAPI-based server for online serving that is compatible with the OpenAI API.
-
-# Interface design
-
-If you use vLLM, then you know how to use vLLM-Omni from Day 0:
-
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: dark)" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/vllm-omni-user-interface.png">
-    <img alt="vLLM-Omni interface design" src="https://raw.githubusercontent.com/vllm-project/vllm-omni/refs/heads/main/docs/source/architecture/vllm-omni-user-interface.png" width=55%>
-  </picture>
-</p>
-
-Taking **Qwen3-Omni** as an example:
-
-## Offline Inference
-The **Omni** class provides a Python interface for offline batched inference. Users initialize the Omni class with a Hugging Face model name and use the generate method, passing inputs that include both text prompts and multi-modal data:
-
-```
-# Create an omni_lm with HF model name.
-from vllm_omni.entrypoints.omni import Omni
-
-omni_lm = Omni(model="Qwen/Qwen3-Omni-30B-A3B-Instruct")
-
-# Example prompts.
-om_inputs = {"prompt": prompt,
-             "multi_modal_data": {
-                 "video": video_frames,
-                 "audio": audio_signal,
-             }}
-
-# Generate texts and audio from the multi-modality inputs.
-outputs = omni_lm.generate(om_inputs, sampling_params_list)
-```
-
-## Online Serving
-Similar to vLLM, vLLM-Omni also provides a FastAPI-based server for online serving. Users can launch the server using the vllm serve command with the `--omni` flag:
-
-```
-vllm serve Qwen/Qwen3-Omni-30B-A3B-Instruct --omni --port 8091
-```
-
-Users can send requests to the server using curl:
-
-```
-# prepare user content
-user_content='[
-        {
-          "type": "video_url",
-          "video_url": {
-            "url": "'"$SAMPLE_VIDEO_URL"'"
-          }
-        },
-        {
-          "type": "text",
-          "text": "Why is this video funny?"
-        }
-      ]'
-    sampling_params_list='[
-      '"$thinker_sampling_params"',
-      '"$talker_sampling_params"',
-      '"$code2wav_sampling_params"'
-    ]'
-    mm_processor_kwargs="{}"
-
-# send the request
-curl -sS -X POST http://localhost:8091/v1/chat/completions \
-    -H "Content-Type: application/json" \
-    -d @- <<EOF
-{
-  "model": "Qwen/Qwen3-Omni-30B-A3B-Instruct",
-  "sampling_params_list": $sampling_params_list,
-  "mm_processor_kwargs": $mm_processor_kwargs,
-  "messages": [
-    {
-      "role": "system",
-      "content": [
-        {
-          "type": "text",
-          "text": "You are Qwen, a virtual human developed by the Qwen Team, Alibaba Group, capable of perceiving auditory and visual inputs, as well as generating text and speech."
-        }
-      ]
-    },
-    {
-      "role": "user",
-      "content": $user_content
-    }
-  ]
-}
-```
-
-For more usages, please refer to [examples](../user_guide/examples/).

docs/design/feature/omni_connector_design.md

@@ -1,200 +0,0 @@
-# OmniConnector: a unified connector interface of disaggregated inference for omni-modality models
-
-This guide explains how to configure and use distributed connectors (vllm_omni/distributed/connectors) in vllm-omni for multi-stage pipelines.
-
-## 1. Overview
-
-Connectors enable data transfer between pipeline stages (e.g., Thinker -> Talker).
-Currently supported connectors operate in **D2H2D (Device-to-Host-to-Device)** mode:
-1. **SharedMemoryConnector**: Uses system shared memory.
-2. **MooncakeConnector**: Uses [Mooncake](https://github.com/kvcache-ai/Mooncake).
-
-*   **SharedMemoryConnector (Default)**: Zero-copy (on host), lowest latency. Best for **single-node** deployments. Auto-configured if no connectors are specified.
-*   **MooncakeConnector**: TCP/RDMA based. Best for **multi-node** distributed deployments. Requires a Mooncake Master service.
-
-## 2. API Design
-
-The connector system is built around the `OmniConnectorBase` abstraction, which decouples data transport from stage logic.
-
-### Core Interface
-
-```python
-class OmniConnectorBase(ABC):
-    @abstractmethod
-    def put(self, from_stage: str, to_stage: str, request_id: str, data: Any) -> tuple[bool, int, Optional[dict]]:
-        """
-        Store data.
-        Returns: (success, serialized_size, metadata)
-        """
-        pass
-
-    @abstractmethod
-    def get(self, from_stage: str, to_stage: str, request_id: str, metadata: Optional[dict] = None) -> Optional[tuple[Any, int]]:
-        """
-        Retrieve data.
-        Args: metadata - Transport-specific handles returned by put() (e.g., SHM name).
-        Returns: (object, serialized_size)
-        """
-        pass
-```
-
-### Key Concept: Metadata Passing
-Unlike a pure key-value store, some connectors (like `SharedMemoryConnector`) generate transient resources (e.g., a shared memory block name) during `put()`. This `metadata` **must be passed via the control plane** (e.g., HTTP headers, queue messages) from the producer stage to the consumer stage so `get()` can locate the data.
-
-## 3. Backends & Use Cases
-
-### 3.1 SharedMemoryConnector
-**Best for:** Single-node, high-performance IPC.
-
-*   **Mechanism:**
-    *   **Small Payloads (< Threshold)**: Data is serialized and passed directly "inline" within the `metadata` dictionary. This avoids the overhead of creating SHM blocks for tiny messages.
-    *   **Large Payloads (>= Threshold)**: Data is written to a named System Shared Memory block. The block name is returned in `metadata`.
-*   **Configuration:**
-    *   `shm_threshold_bytes`: Size in bytes to switch from inline to SHM (default: 64KB).
-
-### 3.2 MooncakeConnector
-**Best for:** Multi-node distributed inference.
-
-*   **Mechanism:** Uses Mooncake's distributed KVCache store.
-    *   **Data Plane**: TCP or RDMA for high-bandwidth transfer.
-    *   **Control Plane**: Uses a centralized Mooncake Master and Metadata Server.
-    *   **Keying**: Deterministic keys based on `request_id/from_stage_to_stage`.
-*   **Requirements**: Requires a running Mooncake Master service.
-
-## 4. Relationship with vLLM
-
-vLLM provides specialized distributed mechanisms for specific artifacts:
-*   **KV Transfer** (`vllm.distributed.kv_transfer`): Optimized for transferring KV caches between prefill and decode instances (using NCCL, Mooncake, etc.).
-*   **EC Transfer** (`vllm.distributed.ec_transfer`): Optimized for sharing encoder embeddings.
-*   **Device Communicators** (`vllm.distributed.device_communicators`): Low-level primitives (NCCL, SHM) for Tensor/Pipeline Parallelism.
-
-`vllm-omni` complements this by introducing a **Generalized Connector Abstraction** (`OmniConnector`) for multimodal pipelines. While vLLM's connectors are artifact-specific, `vllm-omni`:
-
-1.  **Unifies Transport**: Provides a single API (`put`/`get`) to transport *any* stage artifact (Input Embeddings, Hidden States, Audio/Image Tensors, KV Cache, Final Output) between arbitrary pipeline stages (e.g., AudioEncoder -> LLM -> AudioGenerator).
-2.  **Extends Connectivity**: Enables flexible construction of complex DAGs (Directed Acyclic Graphs) where stages can run in the same process, same node, or across nodes, using the most appropriate backend (SHM, Mooncake, etc.) for each edge.
-3.  **Wraps & Adapts**: Can internally utilize vLLM's specialized `kv_transfer` for KV paths while using generic transports (SHM/Mooncake) for other data types, presenting a consistent interface to the application layer.
-
-## 5. Installation (Mooncake)
-
-If using `MooncakeConnector`, install the library first:
-
-```bash
-# For CUDA-enabled systems (Recommended)
-pip install mooncake-transfer-engine
-
-# For non-CUDA systems
-pip install mooncake-transfer-engine-non-cuda
-```
-
-## 6. Using MooncakeConnector
-
-### 6.1 Start Mooncake Master
-
-Start the master service on your primary node:
-
-```bash
-# if you use mooncake SSD storage
-mkdir -p ./mc_storage
-
-mooncake_master \
-  --rpc_port=50051 \
-  --enable_http_metadata_server=true \
-  --http_metadata_server_host=0.0.0.0 \
-  --http_metadata_server_port=8080 \
-  --metrics_port=9003 \
-  --root_fs_dir=./mc_storage/ \
-  --cluster_id=mc-local-1 &
-```
-
-### 6.2 Configuration (YAML)
-
-Edit your stage config (e.g., `qwen2_5_omni.yaml`).
-
-**Step 1: Define Connector in Global Runtime**
-
-```yaml
-runtime:
-  connectors:
-    connector_of_mooncake:
-      name: MooncakeConnector
-      extra:
-        host: "127.0.0.1"           # Local Worker IP
-        metadata_server: "http://<MASTER_IP>:8080/metadata"
-        master: "<MASTER_IP>:50051"
-        segment: 512000000          # 512MB segment
-        localbuf: 64000000          # 64MB buffer
-        proto: "tcp"                # "tcp" or "rdma"
-    ```
-
-**Mooncake Configuration Parameters:**
-
-*   **host**: The hostname or IP address of the local machine (worker). Mooncake uses this to register itself in the metadata server so other nodes can find it.
-*   **metadata_server**: The URL of the metadata server. This is used for service discovery and connection establishment (e.g., exchanging QP information for RDMA).
-*   **master**: The address of the Mooncake Master Server (e.g., `<MASTER_IP>:50051`). This is used for global state management and control plane operations.
-*   **segment**: The size of the global memory segment in bytes (default: ~512MB). This defines the shared memory region accessible by Mooncake for data transfer.
-*   **localbuf**: The size of the local buffer in bytes (default: ~64MB). Used for local data buffering during transfer operations.
-*   **proto**: The transport protocol to use. Options:
-    *   `tcp`: Standard TCP/IP (easier setup, universal compatibility).
-    *   `rdma`: Remote Direct Memory Access (higher performance, requires RDMA-capable hardware).
-
-For more details, refer to the [Mooncake Repository](https://github.com/kvcache-ai/Mooncake).
-
-    **Step 2: Reference in Stages**
-
-Explicitly link stages using `input_connectors` and `output_connectors`:
-
-```yaml
-stage_args:
-  - stage_id: 0
-    # ...
-    output_connectors:
-      to_stage_1: connector_of_mooncake
-
-  - stage_id: 1
-    # ...
-    input_connectors:
-      from_stage_0: connector_of_mooncake
-```
-
-## 7. Using SharedMemoryConnector (Auto-Mode)
-
-**Best for single-node.**
-
-The system will automatically create `SharedMemoryConnector`s for any pipeline edge that does not have an explicit connector defined. This is inferred from:
-1.  `runtime.edges` list in the config.
-2.  `engine_input_source` dependencies defined in `stage_args`.
-
-### Threshold Configuration
-By default, payloads larger than **64KB** (default threshold) are transferred via shared memory, while smaller ones use the control queue (inline).
-
-To adjust this threshold (e.g., to 1GB), add the following to your `runtime.connectors`:
-
-```yaml
-runtime:
-  connectors:
-    connector_of_shared_memory:
-      name: SharedMemoryConnector
-      extra:
-        shm_threshold_bytes: 1024 # 1KB threshold
-```
-
-## 8. Summary
-
-| Use Case | Recommended Connector | Configuration |
-| :--- | :--- | :--- |
-| **Single Node** | `SharedMemoryConnector` | **None** (Automatic) or Custom Threshold |
-| **Multi Node** | `MooncakeConnector` | Explicit YAML + Mooncake Master |
-
-## 9. Operational Notes (important)
-
-- **Fail-fast config validation**: the loader raises if any expected edge is missing a connector. Define `input_connectors`/`output_connectors` or rely on auto-SHM filling; otherwise startup aborts.
-- **Missing payloads halt the stage**: workers expect connector payloads; if metadata or connector config is missing, the stage raises and stops. Verify connector wiring and metadata propagation before production.
-
-## 10. Future Roadmap: Device-to-Device (D2D) Transport
-
-The current implementations (`SharedMemoryConnector`, `MooncakeConnector`) utilize a **D2H2D (Device-to-Host-to-Device)** data path. Tensors are moved to CPU memory (Host) for transport, which incurs PCIe overhead.
-
-As outlined in the design RFC, future versions will introduce **D2D (Device-to-Device)** connectors:
-
-*   **Goal**: Direct GPU-to-GPU transfer (via NCCL, UCX, or IPC) to minimize latency for large tensor payloads.
-*   **Mechanism**: The `OmniConnector` API allows `put()` to initiate a transfer and return a lightweight handle (metadata) via the control plane, while the heavy payload flows directly between devices.

docs/design/feature/ray_based_execution.md

@@ -1,61 +0,0 @@
-# VLLM-Omni Distributed Utilities
-
-This directory (vllm_omni/distributed/ray_utils) contains utilities for distributed execution in vllm-omni, supporting both **Ray** and **Multiprocessing** backends.
-
-## 1. Ray Utils
-
-The `ray_utils` module provides helper functions for managing Ray clusters and actors, which is essential for:
-*   **Multi-node deployment**: Running pipeline stages across different physical machines.
-*   **Resource management**: Efficient GPU/CPU allocation.
-
-### 1.1 Basic Usage
-
-To use the Ray backend, specify `worker_backend="ray"` when initializing the engine.
-
-**Command Line Example:**
-```bash
-vllm serve Qwen/Qwen2.5-Omni-7B \
-  --omni \
-  --port 8091 \
-  --worker-backend ray \
-  --ray-address auto
-```
-
-### 1.2 Cluster Setup
-
-**Step 1: Start Head Node**
-Run this on your primary machine:
-```bash
-ray start --head --port=6399
-```
-
-**Step 2: Connect Worker Nodes**
-Run this on each worker machine:
-```bash
-ray start --address=<HEAD_NODE_IP>:6399
-```
-
-> **Tip**: For a complete cluster setup script, refer to the vLLM example:
-> [run_cluster.sh](https://github.com/vllm-project/vllm/blob/main/examples/online_serving/run_cluster.sh)
-
-### 1.3 Distributed Connector Support
-
-When running on Ray, the system automatically adapts its communication strategy:
-
-*   **Cross-Node**: Recommended to use `MooncakeConnector` (requires separate configuration).
-*   **Same-Node**: Can still use `SharedMemoryConnector` for efficiency, or Ray's native object store (plasma).
-*   **SHM threshold default differs**: when `worker_backend="ray"`, the SharedMemoryConnector default threshold is set to `sys.maxsize`, which forces payloads to go inline (no SHM). Override `shm_threshold_bytes` in the connector config if you want SHM for Ray runs.
-
-### 1.4 Internal Helpers
-
-*   **`initialize_ray_cluster`**: Connects to an existing Ray cluster or starts a local one.
-
-## 2. Troubleshooting
-
-*   **Connection Issues**: Ensure the Ray head node is accessible and ports (default 6399 in this example) are open.
-*   **Version Mismatch**: Ensure all nodes run the same version of Ray and Python.
-
-### Installation
-```bash
-pip install "ray[default]"
-```

docs/design/index.md

@@ -1,19 +0,0 @@
-# Design Documents
-
-This section contains design documents and architecture specifications for vLLM-Omni.
-
-## Architecture Documents
-
-- [Architecture Overview](architecture_overview.md)
-
-## Feature Design Documents
-
-- [Multi-Request Streaming](feature/multi_request_streaming_design.md)
-- [Omni Connector Design](feature/omni_connector_design.md)
-- [Ray-based Execution](feature/ray_based_execution.md)
-
-## Module Design Documents
-
-- [AR Module Design](module/ar_module_design.md)
-- [DIT Module Design](module/dit_module_design.md)
-- [Entrypoint Module Design](module/entrypoint_module_design.md)

docs/design/module/ar_module_design.md

@@ -1 +0,0 @@
-Architecture design of the AR module (update soon)

docs/design/module/dit_module_design.md

@@ -1 +0,0 @@
-Architecture design of the DiT module (update soon)

docs/design/module/entrypoint_module_design.md

@@ -1 +0,0 @@
-Architecture design of the entrypoint (update soon)

docs/examples/README.md

@@ -1,6 +0,0 @@
-# Examples
-
-vLLM-Omni's examples are split into two categories:
-
-- If you are using vLLM-Omni from within Python code, see the *Offline Inference* section.
-- If you are using vLLM-Omni from an HTTP application or client, see the *Online Serving* section.

docs/getting_started/installation/.nav.yml

@@ -1,4 +0,0 @@
-nav:
-  - README.md
-  - gpu.md
-  - npu.md

docs/getting_started/installation/README.md

@@ -1,8 +0,0 @@
-# Installation
-
-vLLM-Omni supports the following hardware platforms:
-
-- [GPU](gpu.md)
-    - [NVIDIA CUDA](gpu.md)
-    - [AMD ROCm](gpu.md)
-- [NPU](npu.md)

docs/getting_started/installation/gpu.md

@@ -1,66 +0,0 @@
-# GPU
-
-vLLM-Omni is a Python library that supports the following GPU variants. The library itself mainly contains python implementations for framework and models.
-
-## Requirements
-
-- OS: Linux
-- Python: 3.12
-
-!!! note
-    vLLM-Omni is currently not natively supported on Windows.
-
-=== "NVIDIA CUDA"
-
-    --8<-- "docs/getting_started/installation/gpu/cuda.inc.md:requirements"
-
-=== "AMD ROCm"
-
-    --8<-- "docs/getting_started/installation/gpu/rocm.inc.md:requirements"
-
-## Set up using Python
-
-### Create a new Python environment
-
---8<-- "docs/getting_started/installation/python_env_setup.inc.md"
-
-### Pre-built wheels
-
-=== "NVIDIA CUDA"
-
-    --8<-- "docs/getting_started/installation/gpu/cuda.inc.md:pre-built-wheels"
-
-
-=== "AMD ROCm"
-
-    --8<-- "docs/getting_started/installation/gpu/rocm.inc.md:pre-built-wheels"
-
-[](){ #build-from-source }
-
-### Build wheel from source
-
-=== "NVIDIA CUDA"
-
-    --8<-- "docs/getting_started/installation/gpu/cuda.inc.md:build-wheel-from-source"
-
-=== "AMD ROCm"
-
-    --8<-- "docs/getting_started/installation/gpu/rocm.inc.md:build-wheel-from-source"
-
-## Set up using Docker
-
-### Build wheel from source
-
-=== "NVIDIA CUDA"
-
-    --8<-- "docs/getting_started/installation/gpu/cuda.inc.md:build-wheel-from-source-in-docker"
-
-=== "AMD ROCm"
-
-    --8<-- "docs/getting_started/installation/gpu/rocm.inc.md:build-wheel-from-source-in-docker"
-
-### Pre-built images
-
-=== "NVIDIA CUDA"
-
-    --8<-- "docs/getting_started/installation/gpu/cuda.inc.md:pre-built-images"

docs/getting_started/installation/gpu/cuda.inc.md

@@ -1,99 +0,0 @@
-# --8<-- [start:requirements]
-
-- GPU: compute capability 7.0 or higher (e.g., V100, T4, RTX20xx, A100, L4, H100, etc.)
-
-# --8<-- [end:requirements]
-# --8<-- [start:set-up-using-python]
-
-vLLM-Omni depends vLLM. So please follow instructions below mainly for vLLM.
-
-!!! note
-    PyTorch installed via `conda` will statically link `NCCL` library, which can cause issues when vLLM tries to use `NCCL`. See <gh-issue:8420> for more details.
-
-In order to be performant, vLLM has to compile many cuda kernels. The compilation unfortunately introduces binary incompatibility with other CUDA versions and PyTorch versions, even for the same PyTorch version with different building configurations.
-
-Therefore, it is recommended to install vLLM and vLLM-Omni with a **fresh new** environment. If either you have a different CUDA version or you want to use an existing PyTorch installation, you need to build vLLM from source. See [build-from-source-vllm](https://docs.vllm.ai/en/stable/getting_started/installation/gpu/#build-wheel-from-source) for more details.
-
-# --8<-- [start:pre-built-wheels]
-
-#### Installation of vLLM
-
-vLLM-Omni is built based on vLLM v0.11.0. Please install it with command below.
-```bash
-uv pip install vllm==0.11.0 --torch-backend=auto
-```
-
-#### Installation of vLLM-Omni
-
-```bash
-uv pip install vllm-omni
-```
-
-# --8<-- [end:pre-built-wheels]
-
-# --8<-- [start:build-wheel-from-source]
-
-#### Installation of vLLM
-If you do not need to modify source code of vLLM, you can directly install the stable 0.11.0 release version of the library
-
-```bash
-uv pip install vllm==0.11.0 --torch-backend=auto
-```
-
-#### Installation of vLLM-Omni
-Install additional requirements for vLLM-Omni
-```bash
-git clone https://github.com/vllm-project/vllm-omni.git
-cd vllm-omni
-uv pip install -e .
-```
-
-<details><summary>(Optional) Installation of vLLM from source</summary>
-If you want to check, modify or debug with source code of vLLM, install the library from source with the following instructions:
-
-```bash
-git clone https://github.com/vllm-project/vllm.git
-cd vllm
-git checkout v0.11.0
-```
-Set up environment variables to get pre-built wheels. If there are internet problems, just download the whl file manually. And set `VLLM_PRECOMPILED_WHEEL_LOCATION` as your local absolute path of whl file.
-```bash
-export VLLM_PRECOMPILED_WHEEL_LOCATION=https://github.com/vllm-project/vllm/releases/download/v0.11.0/vllm-0.11.0-cp38-abi3-manylinux1_x86_64.whl
-```
-Install vllm with command below (If you have no existing PyTorch).
-```bash
-uv pip install --editable .
-```
-Install vllm with command below (If you already have PyTorch).
-```bash
-python use_existing_torch.py
-uv pip install -r requirements/build.txt
-uv pip install --no-build-isolation --editable .
-```
-</details>
-
-# --8<-- [end:build-wheel-from-source]
-
-# --8<-- [start:build-wheel-from-source-in-docker]
-
-# --8<-- [end:build-wheel-from-source-in-docker]
-
-# --8<-- [start:pre-built-images]
-
-vLLM-Omni offers an official docker image for deployment. These images are built on top of vLLM docker images and available on Docker Hub as [vllm/vllm-omni](https://hub.docker.com/r/vllm/vllm-omni/tags). The version of vLLM-Omni indicates which release of vLLM it is based on.
-
-Here's an example deployment command that has been verified on 2 x H100's:
-```bash
-docker run --runtime nvidia --gpus 2 \
-    -v ~/.cache/huggingface:/root/.cache/huggingface \
-    --env "HF_TOKEN=$HF_TOKEN" \
-    -p 8091:8091 \
-    --ipc=host \
-    vllm/vllm-omni:v0.11.0rc1 \
-    --model Qwen/Qwen3-Omni-30B-A3B-Instruct --port 8091
-```
-
-!!! tip
-    You can use this docker image to serve models the same way you would with in vLLM! To do so, make sure you overwrite the default entrypoint (`vllm serve --omni`) which works only for models supported in the vLLM-Omni project.
-
-# --8<-- [end:pre-built-images]

docs/getting_started/installation/gpu/rocm.inc.md

@@ -1,115 +0,0 @@
-# --8<-- [start:requirements]
-
-- GPU: Validated on gfx942 (It should be supported on the AMD GPUs that are supported by vLLM.)
-
-# --8<-- [end:requirements]
-# --8<-- [start:set-up-using-python]
-
-vLLM-Omni current recommends the steps in under setup through Docker Images.
-
-# --8<-- [start:pre-built-wheels]
-
-# --8<-- [end:pre-built-wheels]
-
-# --8<-- [start:build-wheel-from-source]
-
-# --8<-- [end:build-wheel-from-source]
-
-# --8<-- [start:build-wheel-from-source-in-docker]
-
-#### Get the correct docker environment
-
-The v0.11.0 releases is on Oct 3, 2025, so we will pull a AMD nightly docker image close to the release period. `rocm/vllm-dev:nightly_main_20251005` is picked as it has a correct `transformers` version that supports qwen3_omni.
-
-`docker pull rocm/vllm-dev:nightly_main_20251005`
-
-#### Launch the docker environment
-
-In this docker environment, we have all the dependencies for vLLM installed, like `flash-attn` and `aiter`
-
-```bash
-#!/bin/bash
-docker run -it \
-   --privileged \
-   --network=host \
-   --group-add=video \
-   --ipc=host \
-   --cap-add=SYS_PTRACE \
-   --security-opt seccomp=unconfined \
-   --device /dev/kfd \
-   --device /dev/dri \
-   --name vllmomni \
-   rocm/vllm-dev:nightly_main_20251005 \
-   bash
-```
-
-#### Update docker dependencies
-
-```bash
-sudo apt update
-sudo apt install ffmpeg -y
-```
-
-#### Installation of vLLM
-Install the v0.11.0 release of vLLM as vLLM-Omni is currently depend on v0.11.0.
-
-```bash
-# uninstall existing vllm which is based on the main branch, not on the releases branch
-python3 -m pip uninstall -y vllm
-git clone https://github.com/vllm-project/vllm.git vllm0110
-cd vllm0110/
-
-# checkout the version tag v0.11.0
-git checkout -b v0.11.0 v0.11.0
-
-# you should see 0.11.0
-python -c "import setuptools_scm; print(setuptools_scm.get_version())"
-
-# PYTORCH_ROCM_ARCH=<your-gpu-arch> python3 setup.py develop
-PYTORCH_ROCM_ARCH=gfx942 python3 setup.py develop
-```
-
-!!! note
-    vLLM release wheels based on the branch with prefix `releases/`, not from the tag as vLLM may cherry pick bugfixes after cutting a branch.
-
-
-#### Installation of vLLM-Omni
-
-There are two approaches to installing `vllm-omni`
-
-
-1. Install from PyPI
-
-```bash
-pip install vllm-omni
-
-# you might want to also download the source code of the same version to run the examples
-```
-
-2. Install from source
-
-```bash
-# clone and checkout to the release version.
-git clone https://github.com/vllm-project/vllm-omni.git
-cd vllm-omni
-git checkout v0.11.0rc1
-pip install -e .
-```
-
-#### Setup Environment Variables
-
-After installing `vllm` and `vllm-omni` please setup the environment variable as well before running any of the examples and command.
-
-```bash
-export MIOPEN_FIND_MODE=FAST
-export VLLM_ROCM_USE_AITER=1
-export VLLM_ROCM_USE_AITER_MHA=1
-export VLLM_ROCM_USE_AITER_LINEAR=0
-export VLLM_ROCM_USE_AITER_RMSNORM=0
-```
-
-# --8<-- [end:build-wheel-from-source-in-docker]
-
-# --8<-- [start:pre-built-images]
-
-# --8<-- [end:pre-built-images]

docs/getting_started/installation/npu.md

@@ -1,23 +0,0 @@
-# NPU
-
-vLLM-Omni supports NPU through the vLLM Ascend Plugin (vllm-ascend). This is a community maintained hardware plugin for running vLLM on NPU.
-
-## Requirements
-
-- OS: Linux
-- Python: 3.12
-
-!!! note
-    vLLM-Omni is currently not natively supported on Windows.
-
-=== "NPU"
-
-    --8<-- "docs/getting_started/installation/npu/npu.inc.md:requirements"
-
-## Installation
-
-### Recommended
-
-=== "NPU"
-
-    --8<-- "docs/getting_started/installation/npu/npu.inc.md:installation"

docs/getting_started/installation/npu/npu.inc.md

@@ -1,49 +0,0 @@
-# --8<-- [start:requirements]
-
-For detailed hardware and software requirements, please refer to the [vllm-ascend installation documentation](https://docs.vllm.ai/projects/ascend/en/latest/installation.html).
-
-# --8<-- [end:requirements]
-# --8<-- [start:installation]
-
-The recommended way to use vLLM-Omni on NPU is through the vllm-ascend pre-built Docker images:
-
-```bash
-# Update DEVICE according to your NPUs (/dev/davinci[0-7])
-export DEVICE0=/dev/davinci0
-export DEVICE1=/dev/davinci1
-# Update the vllm-ascend image
-# Atlas A2:
-# export IMAGE=quay.io/ascend/vllm-ascend:v0.11.0rc2
-# Atlas A3:
-# export IMAGE=quay.io/ascend/vllm-ascend:v0.11.0rc2-a3
-export IMAGE=quay.io/ascend/vllm-ascend:v0.11.0rc2
-docker run --rm \
-    --name vllm-omni-npu \
-    --shm-size=1g \
-    --device $DEVICE0 \
-    --device $DEVICE1 \
-    --device /dev/davinci_manager \
-    --device /dev/devmm_svm \
-    --device /dev/hisi_hdc \
-    -v /usr/local/dcmi:/usr/local/dcmi \
-    -v /usr/local/bin/npu-smi:/usr/local/bin/npu-smi \
-    -v /usr/local/Ascend/driver/lib64/:/usr/local/Ascend/driver/lib64/ \
-    -v /usr/local/Ascend/driver/version.info:/usr/local/Ascend/driver/version.info \
-    -v /etc/ascend_install.info:/etc/ascend_install.info \
-    -v /root/.cache:/root/.cache \
-    -p 8000:8000 \
-    -it $IMAGE bash
-
-# Inside the container, install vLLM-Omni from source
-cd /vllm-workspace
-git clone https://github.com/vllm-project/vllm-omni.git
-cd vllm-omni
-pip install -v -e .
-export VLLM_WORKER_MULTIPROC_METHOD=spawn
-```
-
-The default workdir is `/workspace`, with vLLM, vLLM-Ascend and vLLM-Omni code placed in `/vllm-workspace` installed in development mode.
-
-For other installation methods (pip installation, building from source, custom Docker builds), please refer to the [vllm-ascend installation guide](https://docs.vllm.ai/projects/ascend/en/latest/installation.html).
-
-# --8<-- [end:installation]

docs/getting_started/installation/python_env_setup.inc.md

@@ -1,6 +0,0 @@
-It's recommended to use [uv](https://docs.astral.sh/uv/), a very fast Python environment manager, to create and manage Python environments. Please follow the [documentation](https://docs.astral.sh/uv/#getting-started) to install `uv`. After installing `uv`, you can create a new Python environment using the following commands:
-
-```bash
-uv venv --python 3.12 --seed
-source .venv/bin/activate
-```

docs/getting_started/quickstart.md

@@ -1,35 +0,0 @@
-# Quickstart
-
-This guide will help you quickly get started with vLLM-Omni to perform:
-
-- Offline batched inference
-- Online serving using OpenAI-compatible server
-
-## Prerequisites
-
-- OS: Linux
-- Python: 3.12
-
-## Installation
-
-Please refer to [installation](installation/README.md)
-
-## Offline Inference
-
-Text-to-image generation quickstart with vLLM-Omni:
-
-```python
-from vllm_omni.entrypoints.omni import Omni
-
-if __name__ == "__main__":
-    omni = Omni(model="Tongyi-MAI/Z-Image-Turbo")
-    prompt = "a cup of coffee on the table"
-    images = omni.generate(prompt)
-    images[0].save("coffee.png")
-```
-
-For more usages, please refer to [offline inference](../user_guide/examples/offline_inference/qwen2_5_omni.md)
-
-## Online Serving with OpenAI-Completions API
-
-Please refer to [online serving](../user_guide/examples/online_serving/qwen2_5_omni.md)

docs/index.md

@@ -0,0 +1,32 @@
+# Welcome to vLLM-omni
+
+**vLLM-omni** extends vLLM capabilities to support multi-modality model inference and serving, particularly focusing on non-autoregressive structures and non-textual outputs. It aims to support:
+
+- **Multi-modal Models**: Text, image, video, audio, and sensor data processing
+- **Non-autoregressive Architectures**: Diffusion Transformers (DiT) and other parallel generation models
+- **Heterogeneous Outputs**: Beyond traditional text generation to multimodal outputs
+
+## <span class="twemoji">🎯</span> Key Features
+
+### Multi-Engine Support
+
+- **Autoregressive Engine**: Traditional text generation with enhanced KV-caching
+- **Diffusion Engine**: Support for DiT models and iterative generation
+- **Hybrid Engine**: Combined AR+DiT processing pipelines
+
+### Modality Processing
+
+- **Text**: Advanced tokenization and embedding generation
+- **Image**: Vision encoder integration (CLIP, etc.)
+- **Audio**: Speech processing and audio embedding
+
+## <span class="twemoji">🤖</span> Supported Models
+
+- Qwen-omni (Thinker-Talker-Codec structure)
+- HunyunaImage 3.0 (Ongoing)
+- Qwen-Image (Ongoing)
+
+## <span class="twemoji">📚</span> Documentation Navigation
+
+- To run open-source models on vLLM-Omni, we recommend starting with the [:material-code-tags: User Quide](user_guide/getting_started/quickstart.md)
+- To develop and contribute to vLLM-Omni, we recommend starting with the [:material-tools: Developer Guide](contributing/README.md)

docs/mkdocs/hooks/generate_api_readme.py

@@ -1,268 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project
-"""
-Hook to automatically generate docs/api/README.md from the codebase.
-
-This script scans the vllm_omni module for public classes and functions,
-categorizes them, and generates a summary README file.
-"""
-
-import ast
-import logging
-from pathlib import Path
-
-logger = logging.getLogger("mkdocs")
-
-ROOT_DIR = Path(__file__).parent.parent.parent.parent
-API_README_PATH = ROOT_DIR / "docs" / "api" / "README.md"
-
-# Category mappings: module prefix -> category name and description
-CATEGORIES = {
-    "entrypoints": {
-        "name": "Entry Points",
-        "description": "Main entry points for vLLM-Omni inference and serving.",
-    },
-    "inputs": {
-        "name": "Inputs",
-        "description": "Input data structures for multi-modal inputs.",
-    },
-    "engine": {
-        "name": "Engine",
-        "description": "Engine classes for offline and online inference.",
-    },
-    "core": {
-        "name": "Core",
-        "description": "Core scheduling and caching components.",
-    },
-    "model_executor": {
-        "name": "Model Executor",
-        "description": "Model execution components.",
-    },
-    "config": {
-        "name": "Configuration",
-        "description": "Configuration classes.",
-    },
-    "worker": {
-        "name": "Workers",
-        "description": "Worker classes and model runners for distributed inference.",
-    },
-}
-
-
-class APIVisitor(ast.NodeVisitor):
-    """AST visitor to extract public classes and module-level functions."""
-
-    def __init__(self, module_path: str):
-        self.module_path = module_path
-        self.classes: list[str] = []
-        self.functions: list[str] = []
-        self._class_stack: list[str] = []  # Track nested class definitions
-
-    def visit_ClassDef(self, node: ast.ClassDef) -> None:
-        """Visit class definitions."""
-        if not node.name.startswith("_"):
-            self.classes.append(f"{self.module_path}.{node.name}")
-        # Track that we're entering a class
-        self._class_stack.append(node.name)
-        self.generic_visit(node)
-        # Remove from stack when done visiting
-        self._class_stack.pop()
-
-    def visit_FunctionDef(self, node: ast.FunctionDef) -> None:
-        """Visit function definitions - only collect module-level functions."""
-        # Only collect if we're not inside a class (stack is empty)
-        if not self._class_stack and not node.name.startswith("_"):
-            self.functions.append(f"{self.module_path}.{node.name}")
-        self.generic_visit(node)
-
-    def visit_AsyncFunctionDef(self, node: ast.AsyncFunctionDef) -> None:
-        """Visit async function definitions - only collect module-level functions."""
-        # Only collect if we're not inside a class (stack is empty)
-        if not self._class_stack and not node.name.startswith("_"):
-            self.functions.append(f"{self.module_path}.{node.name}")
-        self.generic_visit(node)
-
-
-def parse_file_for_symbols(file_path: Path, module_path: str) -> tuple[list[str], list[str]]:
-    """
-    Parse a Python file and extract public classes and functions.
-
-    Returns:
-        Tuple of (classes, functions)
-    """
-    try:
-        # If this is __init__.py, use parent module path
-        if file_path.name == "__init__.py":
-            # Remove .__init__ from module path
-            if module_path.endswith(".__init__"):
-                module_path = module_path[:-9]
-
-        with open(file_path, encoding="utf-8") as f:
-            content = f.read()
-
-        tree = ast.parse(content, filename=str(file_path))
-        visitor = APIVisitor(module_path)
-        visitor.visit(tree)
-
-        return visitor.classes, visitor.functions
-    except Exception as e:
-        logger.debug(f"Could not parse {file_path}: {e}")
-        return [], []
-
-
-def scan_package(package_name: str = "vllm_omni") -> dict[str, list[str]]:
-    """
-    Scan the vllm_omni package and categorize public symbols.
-
-    Returns:
-        Dict mapping category names to lists of symbol full names
-    """
-    categorized: dict[str, list[str]] = {cat["name"]: [] for cat in CATEGORIES.values()}
-
-    try:
-        # Find the package directory
-        package_path = ROOT_DIR / package_name
-        if not package_path.exists():
-            logger.warning(f"Package path not found: {package_path}")
-            return categorized
-
-        # Walk through all Python files
-        for py_file in package_path.rglob("*.py"):
-            # Skip __init__.py and private modules
-            if py_file.name.startswith("_") and py_file.name != "__init__.py":
-                continue
-
-            # Get module path
-            relative_path = py_file.relative_to(ROOT_DIR)
-            module_path = str(relative_path.with_suffix("")).replace("/", ".").replace("\\", ".")
-
-            # Handle __init__.py - use parent module path
-            if py_file.name == "__init__.py":
-                # Remove .__init__ from module path
-                if module_path.endswith(".__init__"):
-                    module_path = module_path[:-9]
-
-            # Determine category from module path
-            category = None
-            for prefix, cat_info in CATEGORIES.items():
-                if prefix in module_path:
-                    category = cat_info["name"]
-                    break
-
-            if not category:
-                continue
-
-            # Parse file for symbols
-            classes, functions = parse_file_for_symbols(py_file, module_path)
-
-            # Filter out internal implementation classes
-            # Skip classes that look like internal components (DiT layers, etc.)
-            internal_patterns = [
-                "Block",
-                "Layer",
-                "Net",
-                "Embedding",
-                "Norm",
-                "Activation",
-                "Solver",
-                "Pooling",
-                "Attention",
-                "MLP",
-                "DecoderLayer",
-                "InputEmbedding",
-                "TimestepEmbedding",
-                "CodecEmbedding",
-                "DownSample",
-                "UpSample",
-                "Res2Net",
-                "SqueezeExcitation",
-                "TimeDelay",
-                "TorchActivation",
-                "SnakeBeta",
-                "SinusPosition",
-                "RungeKutta",
-                "AMPBlock",
-                "AdaLayerNorm",
-            ]
-
-            # Add classes (filter out internal ones)
-            for class_name in classes:
-                class_short_name = class_name.split(".")[-1]
-                # Skip if it matches internal patterns (unless it's a main model class)
-                if any(pattern in class_short_name for pattern in internal_patterns):
-                    # But include main model classes
-                    if not any(
-                        main_class in class_short_name
-                        for main_class in [
-                            "ForConditionalGeneration",
-                            "Model",
-                            "Registry",
-                            "Worker",
-                            "Runner",
-                            "Scheduler",
-                            "Manager",
-                            "Processor",
-                            "Config",
-                        ]
-                    ):
-                        continue
-                categorized[category].append(class_name)
-
-            # Add important functions (parse, preprocess, etc.)
-            for func_name in functions:
-                # Include functions that match certain patterns
-                if any(keyword in func_name.lower() for keyword in ["parse", "preprocess"]):
-                    categorized[category].append(func_name)
-
-        # Sort symbols within each category
-        for category in categorized:
-            categorized[category].sort()
-
-    except Exception as e:
-        logger.error(f"Error scanning package: {e}", exc_info=True)
-
-    return categorized
-
-
-def generate_readme(categorized: dict[str, list[str]]) -> str:
-    """Generate the API README markdown content."""
-    lines = ["# Summary", ""]
-
-    # Generate sections for each category
-    for prefix, cat_info in CATEGORIES.items():
-        category_name = cat_info["name"]
-        description = cat_info["description"]
-        symbols = categorized.get(category_name, [])
-
-        if not symbols:
-            continue
-
-        lines.append(f"## {category_name}")
-        lines.append("")
-        lines.append(description)
-        lines.append("")
-
-        for symbol in symbols:
-            lines.append(f"- [{symbol}][]")
-
-        lines.append("")
-
-    return "\n".join(lines)
-
-
-def on_startup(command, dirty: bool):
-    """MkDocs hook entry point."""
-    logger.info("Generating API README documentation")
-
-    # Scan the package
-    categorized = scan_package()
-
-    # Generate README content
-    content = generate_readme(categorized)
-
-    # Write to file
-    API_README_PATH.parent.mkdir(parents=True, exist_ok=True)
-    with open(API_README_PATH, "w", encoding="utf-8") as f:
-        f.write(content)
-
-    logger.info(f"API README generated: {API_README_PATH.relative_to(ROOT_DIR)}")

docs/mkdocs/hooks/generate_examples.py

@@ -1,5 +1,5 @@
 # SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM-omni project
 import itertools
 import logging
 from dataclasses import dataclass, field
@@ -7,7 +7,6 @@ from pathlib import Path
 from typing import Literal
 
 import regex as re
-import yaml
 
 logger = logging.getLogger("mkdocs")
 
@@ -15,7 +14,6 @@ ROOT_DIR = Path(__file__).parent.parent.parent.parent
 ROOT_DIR_RELATIVE = "../../../../.."
 EXAMPLE_DIR = ROOT_DIR / "examples"
 EXAMPLE_DOC_DIR = ROOT_DIR / "docs/user_guide/examples"
-NAV_FILE = ROOT_DIR / "docs/.nav.yml"
 
 
 def fix_case(text: str) -> str:
@@ -30,6 +28,7 @@ def fix_case(text: str) -> str:
         "lora": "LoRA",
         "rlhf": "RLHF",
         "vllm": "vLLM",
+        "omni": "Omni",
         "openai": "OpenAI",
         "lmcache": "LMCache",
         "multilora": "MultiLoRA",
@@ -103,40 +102,20 @@ class Example:
         """  # noqa: E501
         if self.path.is_file():
             return []
-        # Binary file extensions to exclude
-        binary_extensions = {
-            ".wav",
-            ".mp3",
-            ".mp4",
-            ".avi",
-            ".mov",
-            ".mkv",  # Audio/Video
-            ".png",
-            ".jpg",
-            ".jpeg",
-            ".gif",
-            ".bmp",
-            ".ico",
-            ".svg",  # Images
-            ".pdf",
-            ".zip",
-            ".tar",
-            ".gz",
-            ".bz2",
-            ".xz",  # Archives/Documents
-            ".exe",
-            ".so",
-            ".dll",
-            ".dylib",  # Binaries
-            ".bin",
-            ".dat",
-            ".db",
-            ".sqlite",  # Data files
+        # Text file extensions that can be safely included
+        text_extensions = {
+            ".py",
+            ".md",
+            ".sh",
+            ".yaml",
+            ".yml",
+            ".json",
+            ".txt",
+            ".toml",
+            ".cfg",
+            ".ini",
         }
-
-        def is_other_file(file: Path) -> bool:
-            return file.is_file() and file != self.main_file and file.suffix.lower() not in binary_extensions
-
+        is_other_file = lambda file: (file.is_file() and file != self.main_file and file.suffix in text_extensions)  # noqa: E731
         return [file for file in self.path.rglob("*") if is_other_file(file)]
 
     def determine_title(self) -> str:
@@ -192,10 +171,11 @@ class Example:
         code_fence = "``````"
 
         if self.is_code:
-            main_file_rel = self.main_file.relative_to(ROOT_DIR)
-            content += f'{code_fence}{self.main_file.suffix[1:]}\n--8<-- "{main_file_rel}"\n{code_fence}\n'
+            # Use relative path from ROOT_DIR for snippet inclusion
+            main_file_rel_path = self.main_file.relative_to(ROOT_DIR)
+            content += f'{code_fence}{self.main_file.suffix[1:]}\n--8<-- "{main_file_rel_path}"\n{code_fence}\n'
         else:
-            with open(self.main_file) as f:
+            with open(self.main_file, encoding="utf-8") as f:
                 # Skip the title from md snippets as it's been included above
                 main_content = f.readlines()[1:]
             content += self.fix_relative_links("".join(main_content))
@@ -206,98 +186,18 @@ class Example:
 
         content += "## Example materials\n\n"
         for file in sorted(self.other_files):
+            # Use relative path from ROOT_DIR for snippet inclusion
+            file_rel_path = file.relative_to(ROOT_DIR)
             content += f'??? abstract "{file.relative_to(self.path)}"\n'
             if file.suffix != ".md":
                 content += f"    {code_fence}{file.suffix[1:]}\n"
-            content += f'    --8<-- "{file.relative_to(ROOT_DIR)}"\n'
+            content += f'    --8<-- "{file_rel_path}"\n'
             if file.suffix != ".md":
                 content += f"    {code_fence}\n"
 
         return content
 
 
-def update_nav_file(examples: list[Example]):
-    """
-    Update the .nav.yml file to include all generated examples.
-    This function completely regenerates the examples section based on the actual
-    folder structure, ensuring consistency between the examples folder and nav file.
-
-    Args:
-        examples: List of Example objects that have been generated
-    """
-    if not NAV_FILE.exists():
-        logger.warning("Navigation file not found: %s", NAV_FILE)
-        return
-
-    # Read the current nav file
-    with open(NAV_FILE, encoding="utf-8") as f:
-        nav_data = yaml.safe_load(f) or {}
-
-    nav_list = nav_data.get("nav", [])
-
-    # Find the "User Guide" section
-    user_guide_idx = None
-    examples_idx = None
-    for i, item in enumerate(nav_list):
-        if isinstance(item, dict) and "User Guide" in item:
-            user_guide_idx = i
-            user_guide_content = item["User Guide"]
-            # Find the "Examples" subsection
-            for j, subitem in enumerate(user_guide_content):
-                if isinstance(subitem, dict) and "Examples" in subitem:
-                    examples_idx = j
-                    break
-            break
-
-    if user_guide_idx is None or examples_idx is None:
-        logger.warning("Could not find 'User Guide' -> 'Examples' section in nav file")
-        return
-
-    # Get existing Examples section to preserve non-example items (like README.md)
-    existing_examples_content = nav_list[user_guide_idx]["User Guide"][examples_idx]["Examples"]
-
-    # Preserve string items (like "examples/README.md") that are not example categories
-    preserved_items = [
-        item
-        for item in existing_examples_content
-        if isinstance(item, str) and not item.startswith("user_guide/examples/")
-    ]
-
-    # Group examples by category
-    examples_by_category = {}
-    for example in examples:
-        category = example.category
-        if category not in examples_by_category:
-            examples_by_category[category] = []
-        examples_by_category[category].append(example)
-
-    # Build the new Examples section - start with preserved items
-    examples_section = preserved_items.copy()
-
-    # Add examples grouped by category, sorted by category name
-    for category in sorted(examples_by_category.keys()):
-        category_examples = sorted(examples_by_category[category], key=lambda e: e.path.stem)
-        category_items = []
-        for example in category_examples:
-            doc_path = EXAMPLE_DOC_DIR / example.category / f"{example.path.stem}.md"
-            rel_path = doc_path.relative_to(ROOT_DIR / "docs")
-            category_items.append({example.title: str(rel_path)})
-
-        if category_items:
-            # Format category name (e.g., "offline_inference" -> "Offline Inference")
-            category_title = fix_case(category.replace("_", " ").title())
-            examples_section.append({category_title: category_items})
-
-    # Update the nav structure
-    nav_list[user_guide_idx]["User Guide"][examples_idx]["Examples"] = examples_section
-
-    # Write back to file
-    nav_data["nav"] = nav_list
-    with open(NAV_FILE, "w", encoding="utf-8") as f:
-        yaml.dump(nav_data, f, default_flow_style=False, sort_keys=False, allow_unicode=True)
-    logger.info("Updated navigation file: %s", NAV_FILE.relative_to(ROOT_DIR))
-
-
 def on_startup(command: Literal["build", "gh-deploy", "serve"], dirty: bool):
     logger.info("Generating example documentation")
     logger.debug("Root directory: %s", ROOT_DIR.resolve())
@@ -311,7 +211,7 @@ def on_startup(command: Literal["build", "gh-deploy", "serve"], dirty: bool):
     categories = sorted(p for p in EXAMPLE_DIR.iterdir() if p.is_dir())
 
     examples = []
-    glob_patterns = ["*.py", "*.md", "*.sh"]
+    glob_patterns = ["*.py", "*.md"]
     # Find categorised examples
     for category in categories:
         globs = [category.glob(pattern) for pattern in glob_patterns]
@@ -331,6 +231,3 @@ def on_startup(command: Literal["build", "gh-deploy", "serve"], dirty: bool):
         with open(doc_path, "w+", encoding="utf-8") as f:
             f.write(example.generate())
         logger.debug("Example generated: %s", doc_path.relative_to(ROOT_DIR))
-
-    # Update the navigation file
-    update_nav_file(examples)

docs/mkdocs/hooks/url_schemes.py

@@ -1,121 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""
-This is basically a port of MyST parser’s external URL resolution mechanism
-(https://myst-parser.readthedocs.io/en/latest/syntax/cross-referencing.html#customising-external-url-resolution)
-to work with MkDocs.
-
-It allows Markdown authors to use GitHub shorthand links like:
-
-  - [Text](gh-issue:123)
-  - <gh-pr:456>
-  - [File](gh-file:path/to/file.py#L10)
-
-These are automatically rewritten into fully qualified GitHub URLs pointing to
-issues, pull requests, files, directories, or projects in the
-`vllm-project/vllm-omni` repository.
-
-The goal is to simplify cross-referencing common GitHub resources
-in project docs.
-"""
-
-import regex as re
-from mkdocs.config.defaults import MkDocsConfig
-from mkdocs.structure.files import Files
-from mkdocs.structure.pages import Page
-
-
-def on_page_markdown(markdown: str, *, page: Page, config: MkDocsConfig, files: Files) -> str:
-    """
-    Custom MkDocs plugin hook to rewrite special GitHub reference links
-    in Markdown.
-
-    This function scans the given Markdown content for specially formatted
-    GitHub shorthand links, such as:
-      - `[Link text](gh-issue:123)`
-      - `<gh-pr:456>`
-
-    And rewrites them into fully-qualified GitHub URLs with GitHub icons:
-      - `[:octicons-mark-github-16: Link text](https://github.com/vllm-project/vllm/issues/123)`
-      - `[:octicons-mark-github-16: Pull Request #456](https://github.com/vllm-project/vllm/pull/456)`
-
-    Supported shorthand types:
-      - `gh-issue`
-      - `gh-pr`
-      - `gh-project`
-      - `gh-dir`
-      - `gh-file`
-
-    Args:
-        markdown (str): The raw Markdown content of the page.
-        page (Page): The MkDocs page object being processed.
-        config (MkDocsConfig): The MkDocs site configuration.
-        files (Files): The collection of files in the MkDocs build.
-
-    Returns:
-        str: The updated Markdown content with GitHub shorthand links replaced.
-    """
-    gh_icon = ":octicons-mark-github-16:"
-    gh_url = "https://github.com"
-    repo_url = f"{gh_url}/vllm-project/vllm-omni"
-    org_url = f"{gh_url}/orgs/vllm-project"
-
-    # Mapping of shorthand types to their corresponding GitHub base URLs
-    urls = {
-        "issue": f"{repo_url}/issues",
-        "pr": f"{repo_url}/pull",
-        "project": f"{org_url}/projects",
-        "dir": f"{repo_url}/tree/main",
-        "file": f"{repo_url}/blob/main",
-    }
-
-    # Default title prefixes for auto links
-    titles = {
-        "issue": "Issue #",
-        "pr": "Pull Request #",
-        "project": "Project #",
-        "dir": "",
-        "file": "",
-    }
-
-    # Regular expression to match GitHub shorthand links
-    scheme = r"gh-(?P<type>.+?):(?P<path>.+?)(#(?P<fragment>.+?))?"
-    inline_link = re.compile(r"\[(?P<title>[^\[]+?)\]\(" + scheme + r"\)")
-    auto_link = re.compile(f"<{scheme}>")
-
-    def replace_inline_link(match: re.Match) -> str:
-        """
-        Replaces a matched inline-style GitHub shorthand link
-        with a full Markdown link.
-
-        Example:
-            [My issue](gh-issue:123) → [:octicons-mark-github-16: My issue](https://github.com/vllm-project/vllm/issues/123)
-        """
-        url = f"{urls[match.group('type')]}/{match.group('path')}"
-        if fragment := match.group("fragment"):
-            url += f"#{fragment}"
-
-        return f"[{gh_icon} {match.group('title')}]({url})"
-
-    def replace_auto_link(match: re.Match) -> str:
-        """
-        Replaces a matched autolink-style GitHub shorthand
-        with a full Markdown link.
-
-        Example:
-            <gh-pr:456> → [:octicons-mark-github-16: Pull Request #456](https://github.com/vllm-project/vllm/pull/456)
-        """
-        type = match.group("type")
-        path = match.group("path")
-        title = f"{titles[type]}{path}"
-        url = f"{urls[type]}/{path}"
-        if fragment := match.group("fragment"):
-            url += f"#{fragment}"
-
-        return f"[{gh_icon} {title}]({url})"
-
-    # Replace both inline and autolinks
-    markdown = inline_link.sub(replace_inline_link, markdown)
-    markdown = auto_link.sub(replace_auto_link, markdown)
-
-    return markdown

docs/mkdocs/javascript/edit_and_feedback.js

@@ -1,47 +0,0 @@
-/**
- * edit_and_feedback.js
- *
- * Enhances MkDocs Material docs pages by:
- *
- * 1. Adding a "Question? Give us feedback" link
- *    below the "Edit" button.
- *
- *    - The link opens a GitHub issue with a template,
- *      auto-filled with the current page URL and path.
- *
- * 2. Ensuring the edit button opens in a new tab
- *    with target="_blank" and rel="noopener".
- */
-document.addEventListener("DOMContentLoaded", function () {
-    const url = window.location.href;
-    const page = document.body.dataset.mdUrl || location.pathname;
-
-    const feedbackLink = document.createElement("a");
-    feedbackLink.href = `https://github.com/vllm-project/vllm-omni/issues/new?template=100-documentation.yml&title=${encodeURIComponent(
-      `[Docs] Feedback for \`${page}\``
-    )}&body=${encodeURIComponent(`📄 **Reference:**\n${url}\n\n📝 **Feedback:**\n_Your response_`)}`;
-    feedbackLink.target = "_blank";
-    feedbackLink.rel = "noopener";
-    feedbackLink.title = "Provide feedback";
-    feedbackLink.className = "md-content__button";
-    feedbackLink.innerHTML = `
-    <svg
-      xmlns="http://www.w3.org/2000/svg"
-      height="24px"
-      viewBox="0 -960 960 960"
-      width="24px"
-      fill="currentColor"
-    >
-      <path d="M280-280h280v-80H280v80Zm0-160h400v-80H280v80Zm0-160h400v-80H280v80Zm-80 480q-33 0-56.5-23.5T120-200v-560q0-33 23.5-56.5T200-840h560q33 0 56.5 23.5T840-760v560q0 33-23.5 56.5T760-120H200Zm0-80h560v-560H200v560Zm0-560v560-560Z"/>
-    </svg>
-  `;
-
-    const editButton = document.querySelector('.md-content__button[href*="edit"]');
-
-    if (editButton && editButton.parentNode) {
-      editButton.insertAdjacentElement("beforebegin", feedbackLink);
-
-      editButton.setAttribute("target", "_blank");
-      editButton.setAttribute("rel", "noopener");
-    }
-  });

docs/mkdocs/javascript/mathjax.js

@@ -1,20 +0,0 @@
-// Enables MathJax rendering
-window.MathJax = {
-    tex: {
-      inlineMath: [["\\(", "\\)"]],
-      displayMath: [["\\[", "\\]"]],
-      processEscapes: true,
-      processEnvironments: true
-    },
-    options: {
-      ignoreHtmlClass: ".*|",
-      processHtmlClass: "arithmatex"
-    }
-  };
-
-  document$.subscribe(() => {
-    MathJax.startup.output.clearCache()
-    MathJax.typesetClear()
-    MathJax.texReset()
-    MathJax.typesetPromise()
-  })

docs/mkdocs/javascript/slack_and_forum.js

@@ -1,56 +0,0 @@
-/**
- * slack_and_forum.js
- *
- * Adds a custom Slack and Forum button to the MkDocs Material header.
- *
- */
-
-window.addEventListener('DOMContentLoaded', () => {
-    const headerInner = document.querySelector('.md-header__inner');
-
-    if (headerInner) {
-      const slackButton = document.createElement('button');
-      slackButton.className = 'slack-button';
-      slackButton.title = 'Join us on Slack';
-      slackButton.style.border = 'none';
-      slackButton.style.background = 'transparent';
-      slackButton.style.cursor = 'pointer';
-
-      slackButton.innerHTML = `
-        <img src="https://a.slack-edge.com/80588/marketing/img/icons/icon_slack_hash_colored.png"
-             style="height: 1.1rem;"
-             alt="Slack">
-      `;
-
-      slackButton.addEventListener('click', () => {
-        window.open('https://slack.vllm.ai', '_blank', 'noopener');
-      });
-
-      const forumButton = document.createElement('button');
-      forumButton.className = 'forum-button';
-      forumButton.title = 'Join the Forum';
-      forumButton.style.border = 'none';
-      forumButton.style.background = 'transparent';
-      forumButton.style.cursor = 'pointer';
-
-      forumButton.innerHTML = `
-        <svg
-          xmlns="http://www.w3.org/2000/svg"
-          viewBox="0 -960 960 960"
-          fill="currentColor"
-        >
-          <path d="M817.85-198.15 698.46-317.54H320q-24.48 0-41.47-16.99T261.54-376v-11.69h424.61q25.39 0 43.47-18.08 18.07-18.08 18.07-43.46v-268.92h11.69q24.48 0 41.47 16.99 17 16.99 17 41.47v461.54ZM179.08-434.69l66.84-66.85h363.31q10.77 0 17.69-6.92 6.93-6.92 6.93-17.69v-246.77q0-10.77-6.93-17.7-6.92-6.92-17.69-6.92H203.69q-10.77 0-17.69 6.92-6.92 6.93-6.92 17.7v338.23Zm-36.93 89.46v-427.69q0-25.39 18.08-43.46 18.08-18.08 43.46-18.08h405.54q25.39 0 43.46 18.08 18.08 18.07 18.08 43.46v246.77q0 25.38-18.08 43.46-18.07 18.07-43.46 18.07H261.54L142.15-345.23Zm36.93-180.92V-797.54v271.39Z"/>
-        </svg>
-      `;
-
-      forumButton.addEventListener('click', () => {
-        window.open('https://discuss.vllm.ai/', '_blank', 'noopener');
-      });
-
-      const githubSource = document.querySelector('.md-header__source');
-      if (githubSource) {
-        githubSource.parentNode.insertBefore(slackButton, githubSource.nextSibling);
-        githubSource.parentNode.insertBefore(forumButton, slackButton.nextSibling);
-      }
-    }
-  });

docs/mkdocs/overrides/main.html

@@ -1 +0,0 @@
-{% extends "base.html" %}

docs/mkdocs/overrides/partials/toc-item.html

@@ -1,21 +0,0 @@
-<!-- Enables the use of toc_depth in document frontmatter https://github.com/squidfunk/mkdocs-material/issues/4827#issuecomment-1869812019 -->
-<li class="md-nav__item">
-    <a href="{{ toc_item.url }}" class="md-nav__link">
-      <span class="md-ellipsis">
-        {{ toc_item.title }}
-      </span>
-    </a>
-
-    <!-- Table of contents list -->
-    {% if toc_item.children %}
-      <nav class="md-nav" aria-label="{{ toc_item.title | striptags }}">
-        <ul class="md-nav__list">
-          {% for toc_item in toc_item.children %}
-          {% if not page.meta.toc_depth or toc_item.level <= page.meta.toc_depth %}
-            {% include "partials/toc-item.html" %}
-          {% endif %}
-          {% endfor %}
-        </ul>
-      </nav>
-    {% endif %}
-  </li>

docs/mkdocs/stylesheets/extra.css

@@ -1,221 +1,55 @@
-/* Warning for latest docs */
-.md-banner {
-    background-color: var(--md-warning-bg-color);
-    color: var(--md-warning-fg-color);
+/* Custom CSS for vLLM-omni documentation - based on vLLM's extra.css */
+
+/* Improve code block readability */
+.highlight {
+    border-radius: 4px;
 }
 
-/* https://christianoliff.com/blog/styling-external-links-with-an-icon-in-css/ */
-a:not(:has(svg)):not(.md-icon):not(.autorefs-external) {
-    align-items: center;
-
-    &[href^="//"]::after,
-    &[href^="http://"]::after,
-    &[href^="https://"]::after {
-        content: "";
-        width: 12px;
-        height: 12px;
-        margin-left: 4px;
-        background-image: url("data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' width='16' height='16' stroke='gray' viewBox='0 0 16 16'%3E%3Cpath fill-rule='evenodd' d='M8.636 3.5a.5.5 0 0 0-.5-.5H1.5A1.5 1.5 0 0 0 0 4.5v10A1.5 1.5 0 0 0 1.5 16h10a1.5 1.5 0 0 0 1.5-1.5V7.864a.5.5 0 0 0-1 0V14.5a.5.5 0 0 1-.5.5h-10a.5.5 0 0 1-.5-.5v-10a.5.5 0 0 1 .5-.5h6.636a.5.5 0 0 0 .5-.5z'/%3E%3Cpath fill-rule='evenodd' d='M16 .5a.5.5 0 0 0-.5-.5h-5a.5.5 0 0 0 0 1h3.793L6.146 9.146a.5.5 0 1 0 .708.708L15 1.707V5.5a.5.5 0 0 0 1 0v-5z'/%3E%3C/svg%3E");
-        background-position: center;
-        background-repeat: no-repeat;
-        background-size: contain;
-        display: inline-block;
-    }
+/* Style API documentation */
+.md-typeset .doc-contents {
+    margin-bottom: 1.5em;
 }
 
-a[href*="localhost"]::after,
-a[href*="127.0.0.1"]::after,
-
-/* Hide external link icons for all links */
-a[href^="//"]::after,
-a[href^="http://"]::after,
-a[href^="https://"]::after {
-    display: none !important;
+/* Improve table readability */
+.md-typeset table {
+    display: table;
+    width: 100%;
 }
 
-/* Light mode: darker section titles */
+/* Light mode: darker section titles (from vLLM) */
 body[data-md-color-scheme="default"] .md-nav__item--section > label.md-nav__link .md-ellipsis {
   color: rgba(0, 0, 0, 0.7) !important;
   font-weight: 700;
 }
 
-/* Dark mode: lighter gray section titles */
+/* Dark mode: lighter gray section titles (from vLLM) */
 body[data-md-color-scheme="slate"] .md-nav__item--section > label.md-nav__link .md-ellipsis {
   color: rgba(255, 255, 255, 0.75) !important;
   font-weight: 700;
 }
 
-/* Custom admonitions */
-:root {
-  --md-admonition-icon--announcement: url('data:image/svg+xml;charset=utf-8,<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 16 16" width="16" height="16"><path d="M3.25 9a.75.75 0 0 1 .75.75c0 2.142.456 3.828.733 4.653a.122.122 0 0 0 .05.064.212.212 0 0 0 .117.033h1.31c.085 0 .18-.042.258-.152a.45.45 0 0 0 .075-.366A16.743 16.743 0 0 1 6 9.75a.75.75 0 0 1 1.5 0c0 1.588.25 2.926.494 3.85.293 1.113-.504 2.4-1.783 2.4H4.9c-.686 0-1.35-.41-1.589-1.12A16.4 16.4 0 0 1 2.5 9.75.75.75 0 0 1 3.25 9Z"></path><path d="M0 6a4 4 0 0 1 4-4h2.75a.75.75 0 0 1 .75.75v6.5a.75.75 0 0 1-.75.75H4a4 4 0 0 1-4-4Zm4-2.5a2.5 2.5 0 1 0 0 5h2v-5Z"></path><path d="M15.59.082A.75.75 0 0 1 16 .75v10.5a.75.75 0 0 1-1.189.608l-.002-.001h.001l-.014-.01a5.775 5.775 0 0 0-.422-.25 10.63 10.63 0 0 0-1.469-.64C11.576 10.484 9.536 10 6.75 10a.75.75 0 0 1 0-1.5c2.964 0 5.174.516 6.658 1.043.423.151.787.302 1.092.443V2.014c-.305.14-.669.292-1.092.443C11.924 2.984 9.713 3.5 6.75 3.5a.75.75 0 0 1 0-1.5c2.786 0 4.826-.484 6.155-.957.665-.236 1.154-.47 1.47-.64.144-.077.284-.161.421-.25l.014-.01a.75.75 0 0 1 .78-.061Z"></path></svg>');
-  --md-admonition-icon--important: url('data:image/svg+xml;charset=utf-8,<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 16 16" width="16" height="16"><path d="M4.47.22A.749.749 0 0 1 5 0h6c.199 0 .389.079.53.22l4.25 4.25c.141.14.22.331.22.53v6a.749.749 0 0 1-.22.53l-4.25 4.25A.749.749 0 0 1 11 16H5a.749.749 0 0 1-.53-.22L.22 11.53A.749.749 0 0 1 0 11V5c0-.199.079-.389.22-.53Zm.84 1.28L1.5 5.31v5.38l3.81 3.81h5.38l3.81-3.81V5.31L10.69 1.5ZM8 4a.75.75 0 0 1 .75.75v3.5a.75.75 0 0 1-1.5 0v-3.5A.75.75 0 0 1 8 4Zm0 8a1 1 0 1 1 0-2 1 1 0 0 1 0 2Z"></path></svg>');
-  --md-admonition-icon--code: url('data:image/svg+xml;charset=utf-8,<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 16 16"><path d="m11.28 3.22 4.25 4.25a.75.75 0 0 1 0 1.06l-4.25 4.25a.749.749 0 0 1-1.275-.326.75.75 0 0 1 .215-.734L13.94 8l-3.72-3.72a.749.749 0 0 1 .326-1.275.75.75 0 0 1 .734.215m-6.56 0a.75.75 0 0 1 1.042.018.75.75 0 0 1 .018 1.042L2.06 8l3.72 3.72a.749.749 0 0 1-.326 1.275.75.75 0 0 1-.734-.215L.47 8.53a.75.75 0 0 1 0-1.06Z"/></svg>');
-  --md-admonition-icon--console: url('data:image/svg+xml;charset=utf-8,<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 16 16"><path d="M0 2.75C0 1.784.784 1 1.75 1h12.5c.966 0 1.75.784 1.75 1.75v10.5A1.75 1.75 0 0 1 14.25 15H1.75A1.75 1.75 0 0 1 0 13.25Zm1.75-.25a.25.25 0 0 0-.25.25v10.5c0 .138.112.25.25.25h12.5a.25.25 0 0 0 .25-.25V2.75a.25.25 0 0 0-.25-.25ZM7.25 8a.75.75 0 0 1-.22.53l-2.25 2.25a.749.749 0 0 1-1.275-.326.75.75 0 0 1 .215-.734L5.44 8 3.72 6.28a.749.749 0 0 1 .326-1.275.75.75 0 0 1 .734.215l2.25 2.25c.141.14.22.331.22.53m1.5 1.5h3a.75.75 0 0 1 0 1.5h-3a.75.75 0 0 1 0-1.5"/></svg>');
+/* Navigation: Expand first level of API Reference by default, collapse deeper levels */
+/* Show direct children of API Reference section (vllm_omni's children: config, core, entrypoints, etc.) */
+.md-nav__item--nested[data-md-state] > .md-nav {
+    display: none;
 }
 
-.md-typeset .admonition.announcement,
-.md-typeset details.announcement {
-  border-color: rgb(255, 110, 66);
-}
-.md-typeset .admonition.important,
-.md-typeset details.important {
-  border-color: rgb(239, 85, 82);
-}
-.md-typeset .admonition.code,
-.md-typeset details.code {
-  border-color: #64dd17
-}
-.md-typeset .admonition.console,
-.md-typeset details.console {
-  border-color: #64dd17
+/* Expand active items */
+.md-nav__item--nested.md-nav__item--active > .md-nav {
+    display: block;
 }
 
-.md-typeset .announcement > .admonition-title,
-.md-typeset .announcement > summary {
-  background-color: rgb(255, 110, 66, 0.1);
-}
-.md-typeset .important > .admonition-title,
-.md-typeset .important > summary {
-  background-color: rgb(239, 85, 82, 0.1);
-}
-.md-typeset .code > .admonition-title,
-.md-typeset .code > summary {
-  background-color: #64dd171a;
-}
-.md-typeset .console > .admonition-title,
-.md-typeset .console > summary {
-  background-color: #64dd171a;
+/* For API Reference: show first level children (config, core, entrypoints, etc.) by default */
+.md-nav__item--section:has([href*="api/vllm_omni/index"]) > .md-nav > .md-nav__list > .md-nav__item--nested > .md-nav {
+    display: block;
 }
 
-.md-typeset .announcement > .admonition-title::before,
-.md-typeset .announcement > summary::before {
-  background-color: rgb(239, 85, 82);
-  -webkit-mask-image: var(--md-admonition-icon--announcement);
-          mask-image: var(--md-admonition-icon--announcement);
-}
-.md-typeset .important > .admonition-title::before,
-.md-typeset .important > summary::before {
-  background-color: rgb(239, 85, 82);
-  -webkit-mask-image: var(--md-admonition-icon--important);
-          mask-image: var(--md-admonition-icon--important);
-}
-.md-typeset .code > .admonition-title::before,
-.md-typeset .code > summary::before {
-  background-color: #64dd17;
-  -webkit-mask-image: var(--md-admonition-icon--code);
-          mask-image: var(--md-admonition-icon--code);
-}
-.md-typeset .console > .admonition-title::before,
-.md-typeset .console > summary::before {
-  background-color: #64dd17;
-  -webkit-mask-image: var(--md-admonition-icon--console);
-          mask-image: var(--md-admonition-icon--console);
+/* But keep deeper levels collapsed */
+.md-nav__item--section:has([href*="api/vllm_omni/index"]) > .md-nav > .md-nav__list > .md-nav__item--nested > .md-nav > .md-nav__list > .md-nav__item--nested > .md-nav {
+    display: none;
 }
 
-/* Make label fully visible on hover */
-.md-content__button[href*="edit"]:hover::after {
-  opacity: 1;
-}
-
-/* Hide edit button on generated docs/examples pages */
-@media (min-width: 960px) {
-  .md-content__button[href*="docs/examples/"] {
-    display: none !important;
-  }
-}
-
-.md-content__button-wrapper {
-  position: absolute;
-  top: 0.6rem;
-  right: 0.8rem;
-  display: flex;
-  flex-direction: row;
-  align-items: center;
-  gap: 0.4rem;
-  z-index: 1;
-}
-
-.md-content__button-wrapper a {
-  display: inline-flex;
-  align-items: center;
-  justify-content: center;
-  height: 24px;
-  width: 24px;
-  color: var(--md-default-fg-color);
-  text-decoration: none;
-}
-
-.md-content__button-wrapper a:hover {
-  color: var(--md-accent-fg-color);
-}
-
-/* Slack and Forum css */
-.slack-button,
-.forum-button {
-  display: inline-flex;
-  align-items: center;
-  justify-content: center;
-  margin-left: 0.4rem;
-  height: 24px;
-}
-
-.slack-button img {
-  height: 18px;
-  filter: none !important;
-}
-
-.slack-button:hover,
-.forum-button:hover {
-  opacity: 0.7;
-}
-
-.forum-button svg {
-  height: 28px;
-  opacity: 0.9;
-  transform: translateY(2px);
-}
-
-/* For logo css */
-[data-md-color-scheme="default"] .logo-dark {
-  display: none;
-}
-
-[data-md-color-scheme="slate"] .logo-light {
-  display: none;
-}
-
-/* Outline for content tabs */
-.md-typeset .tabbed-set {
-  border: 0.075rem solid var(--md-default-fg-color);
-  border-radius: 0.2rem;
-}
-
-.md-typeset .tabbed-content {
-  padding: 0 0.6em;
-}
-
-/* Hide link icon in header logo */
-.md-header__button.md-logo :is(img, svg) {
-  pointer-events: none;
-}
-
-.md-header__button.md-logo::after {
-  display: none !important;
-}
-
-/* Hide link icons in content tabs (tabbed content) */
-.md-typeset .tabbed-set > label::after,
-.md-typeset .tabbed-set > input:checked + label::after {
-  display: none !important;
-}
-
-/* Hide link icons in navigation tabs */
-.md-nav__link[href]::after,
-.md-nav__item--nested > .md-nav__link::after {
-  display: none !important;
-}
-
-/* Hide link icons in top navigation tabs */
-.md-tabs__link::after {
-  display: none !important;
+/* Expand deeper levels only when active */
+.md-nav__item--section:has([href*="api/vllm_omni/index"]) > .md-nav > .md-nav__list > .md-nav__item--nested > .md-nav > .md-nav__list > .md-nav__item--nested.md-nav__item--active > .md-nav {
+    display: block;
 }

docs/models/supported_models.md

@@ -1,40 +0,0 @@
-# Supported Models
-
-vLLM-Omni supports unified multimodal comprehension and generation models across various tasks.
-
-## Model Implementation
-
-If vLLM-Omni natively supports a model, its implementation can be found in <gh-file:vllm_omni/model_executor/models> and <gh-file:vllm_omni/diffusion/models>.
-
-## List of Supported Models for Nvidia GPU / AMD GPU
-
-<style>
-th {
-  white-space: nowrap;
-  min-width: 0 !important;
-}
-</style>
-
-| Architecture | Models | Example HF Models |
-|--------------|--------|-------------------|
-| `Qwen3OmniMoeForConditionalGeneration` | Qwen3-Omni | `Qwen/Qwen3-Omni-30B-A3B-Instruct` |
-| `Qwen2_5OmniForConditionalGeneration` | Qwen2.5-Omni | `Qwen/Qwen2.5-Omni-7B`, `Qwen/Qwen2.5-Omni-3B` |
-| `QwenImagePipeline` | Qwen-Image | `Qwen/Qwen-Image` |
-| `QwenImageEditPipeline` | Qwen-Image-Edit | `Qwen/Qwen-Image-Edit` |
-|`ZImagePipeline` | Z-Image | `Tongyi-MAI/Z-Image-Turbo` |
-| `WanPipeline` | Wan2.2 | `Wan-AI/Wan2.2-T2V-A14B-Diffusers` |
-
-
-## List of Supported Models for NPU
-
-<style>
-th {
-  white-space: nowrap;
-  min-width: 0 !important;
-}
-</style>
-
-| Architecture | Models | Example HF Models |
-|--------------|--------|-------------------|
-| `Qwen2_5OmniForConditionalGeneration` | Qwen2.5-Omni | `Qwen/Qwen2.5-Omni-7B`, `Qwen/Qwen2.5-Omni-3B`|
-| `QwenImagePipeline` | Qwen-Image | `Qwen/Qwen-Image` |

docs/usage/faq.md

@@ -1,25 +0,0 @@
-# Frequently Asked Questions
-
-> Q: How many chips do I need to infer a model in vLLM-Omni?
-
-A: Now, we support natively disaggregated deployment for different model stages within a model. There is a restriction that one chip can only have one AutoRegressive model stage. This is because the unified KV cache management of vLLM. Stages of other types can coexist within a chip. The restriction will be resolved in later version.
-
-> Q: When trying to run examples, I encounter error about backend of librosa or soundfile. How to solve it?
-
-A: If you encounter error about backend of librosa, try to install ffmpeg with command below.
-```
-sudo apt update
-sudo apt install ffmpeg
-```
-
-> Q: I encounter some bugs or CI problems, which is urgent. How can I solve it?
-
-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: vLLM-Omni partitions model into several stages. For AR stages, it will reuse main logic of LLMEngine in vLLM. So current quantization supported in vLLM should be also supported in vLLM-Omni for them. But systematic verification is ongoing. For quantization for DiffusionEngine, we are working on it. Please stay tuned and welcome contribution!
-
-> Q: Does vLLM-Omni support multimodal streaming input and output?
-
-A: Not yet. We already put it on the [Roadmap](https://github.com/vllm-project/vllm-omni/issues/165). Please stay tuned!

docs/user_guide/examples/index.md

@@ -0,0 +1,8 @@
+# Examples
+
+vLLM-omni's examples are split into two categories:
+
+- If you are using vLLM-omni from within Python code, see the *Offline Inference* section.
+- If you are using vLLM-omni from an HTTP application or client, see the *Online Serving* section.
+
+For detailed example documentation, check the [examples directory](https://github.com/vllm-project/vllm-omni/tree/main/examples) in the repository.

docs/user_guide/examples/offline_inference/image_to_image.md

@@ -1,30 +0,0 @@
-# Qwen-Image-Edit Offline Inference
-
-Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/offline_inference/image_to_image>. Added in PR [#196](https://github.com/vllm-project/vllm-omni/pull/196).
-
-This example edits an input image with `Qwen/Qwen-Image-Edit` using the `image_edit.py` CLI.
-
-## Local CLI Usage
-
-```bash
-python image_edit.py \
-  --image qwen_bear.png \
-  --prompt "Let this mascot dance under the moon, surrounded by floating stars and poetic bubbles such as 'Be Kind'" \
-  --output output_image_edit.png \
-  --num_inference_steps 50 \
-  --cfg_scale 4.0
-```
-
-Key arguments:
-- `--image`: path to the source image (PNG/JPG, converted to RGB).
-- `--prompt` / `--negative_prompt`: text description (string).
-- `--cfg_scale`: true CFG scale for Qwen-Image-Edit (quality vs. fidelity).
-- `--num_inference_steps`: diffusion sampling steps (more steps = higher quality, slower).
-- `--output`: path to save the generated PNG.
-
-## Example materials
-
-??? abstract "image_edit.py"
-    ``````py
-    --8<-- "examples/offline_inference/image_to_image/image_edit.py"
-    ``````

docs/user_guide/examples/offline_inference/qwen2_5_omni.md

@@ -1,14 +1,13 @@
-# Offline Example of vLLM-Omni for Qwen2.5-Omni
+# Offline Example of vLLM-omni for Qwen2.5-omni
 
-Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/offline_inference/qwen2_5_omni>.
+Source <https://github.com/vllm-project/vllm-omni/tree/main/examples\offline_inference\qwen2_5_omni>.
 
 
 ## 🛠️ Installation
 
 Please refer to [README.md](https://github.com/vllm-project/vllm-omni/tree/main/README.md)
 
-## Run examples (Qwen2.5-Omni)
-
+## Run examples (Qwen2.5-omni)
 ### Multiple Prompts
 Download dataset from [seed_tts](https://drive.google.com/file/d/1GlSjVfSHkW3-leKKBlfrjuuTGqQ_xaLP/edit). To get the prompt, you can:
 ```bash
@@ -27,8 +26,7 @@ Then run the command below.
 ```bash
 bash run_multiple_prompts.sh
 ```
-
-### Single Prompt
+### Single Prompts
 Get into the example folder
 ```bash
 cd examples/offline_inference/qwen2_5_omni
@@ -38,57 +36,29 @@ Then run the command below.
 bash run_single_prompt.sh
 ```
 
-#### Using Local Media Files
-The `end2end.py` script supports local media files (audio, video, image) via CLI arguments:
-
-```bash
-# Use single local media files
-python end2end.py --query-type use_image --image-path /path/to/image.jpg
-python end2end.py --query-type use_video --video-path /path/to/video.mp4
-python end2end.py --query-type use_audio --audio-path /path/to/audio.wav
-
-# Combine multiple local media files
-python end2end.py --query-type mixed_modalities \
-    --video-path /path/to/video.mp4 \
-    --image-path /path/to/image.jpg \
-    --audio-path /path/to/audio.wav
-
-# Use audio from video file
-python end2end.py --query-type use_audio_in_video --video-path /path/to/video.mp4
-
-```
-
-If media file paths are not provided, the script will use default assets. Supported query types:
-- `use_image`: Image input only
-- `use_video`: Video input only
-- `use_audio`: Audio input only
-- `mixed_modalities`: Audio + image + video
-- `use_audio_in_video`: Extract audio from video
-- `text`: Text-only query
-
-### FAQ
-
-If you encounter error about backend of librosa, try to install ffmpeg with command below.
-```
-sudo apt update
-sudo apt install ffmpeg
-```
-
 ## Example materials
 
 ??? abstract "end2end.py"
     ``````py
-    --8<-- "examples/offline_inference/qwen2_5_omni/end2end.py"
+    --8<-- "examples\offline_inference\qwen2_5_omni\end2end.py"
     ``````
 ??? abstract "extract_prompts.py"
     ``````py
-    --8<-- "examples/offline_inference/qwen2_5_omni/extract_prompts.py"
+    --8<-- "examples\offline_inference\qwen2_5_omni\extract_prompts.py"
+    ``````
+??? abstract "processing_omni.py"
+    ``````py
+    --8<-- "examples\offline_inference\qwen2_5_omni\processing_omni.py"
     ``````
 ??? abstract "run_multiple_prompts.sh"
     ``````sh
-    --8<-- "examples/offline_inference/qwen2_5_omni/run_multiple_prompts.sh"
+    --8<-- "examples\offline_inference\qwen2_5_omni\run_multiple_prompts.sh"
     ``````
 ??? abstract "run_single_prompt.sh"
     ``````sh
-    --8<-- "examples/offline_inference/qwen2_5_omni/run_single_prompt.sh"
+    --8<-- "examples\offline_inference\qwen2_5_omni\run_single_prompt.sh"
+    ``````
+??? abstract "utils.py"
+    ``````py
+    --8<-- "examples\offline_inference\qwen2_5_omni\utils.py"
     ``````

docs/user_guide/examples/offline_inference/qwen3_omni.md

@@ -1,81 +0,0 @@
-# Offline Example of vLLM-Omni for Qwen3-Omni
-
-Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/offline_inference/qwen3_omni>.
-
-
-## 🛠️ Installation
-
-Please refer to [README.md](https://github.com/vllm-project/vllm-omni/tree/main/README.md)
-
-## Run examples (Qwen3-Omni)
-
-### Multiple Prompts
-Download dataset from [seed_tts](https://drive.google.com/file/d/1GlSjVfSHkW3-leKKBlfrjuuTGqQ_xaLP/edit). For processing dataset please refer to [Qwen2.5-Omni README.md](https://github.com/vllm-project/vllm-omni/tree/main/examples/offline_inference/qwen2_5_omni/README.md)
-Get into the example folder
-```bash
-cd examples/offline_inference/qwen3_omni
-```
-Then run the command below.
-```bash
-bash run_multiple_prompts.sh
-```
-### Single Prompt
-Get into the example folder
-```bash
-cd examples/offline_inference/qwen3_omni
-```
-Then run the command below.
-```bash
-bash run_single_prompt.sh
-```
-If you have not enough memory, you can set thinker with tensor parallel. Just run the command below.
-```bash
-bash run_single_prompt_tp.sh
-```
-
-#### Using Local Media Files
-The `end2end.py` script supports local media files (audio, video, image) via command-line arguments:
-
-```bash
-# Use local video file
-python end2end.py --query-type use_video --video-path /path/to/video.mp4
-
-# Use local image file
-python end2end.py --query-type use_image --image-path /path/to/image.jpg
-
-# Use local audio file
-python end2end.py --query-type use_audio --audio-path /path/to/audio.wav
-```
-
-If media file paths are not provided, the script will use default assets. Supported query types:
-- `use_video`: Video input
-- `use_image`: Image input
-- `use_audio`: Audio input
-- `text`: Text-only query
-
-### FAQ
-
-If you encounter error about backend of librosa, try to install ffmpeg with command below.
-```
-sudo apt update
-sudo apt install ffmpeg
-```
-
-## Example materials
-
-??? abstract "end2end.py"
-    ``````py
-    --8<-- "examples/offline_inference/qwen3_omni/end2end.py"
-    ``````
-??? abstract "run_multiple_prompts.sh"
-    ``````sh
-    --8<-- "examples/offline_inference/qwen3_omni/run_multiple_prompts.sh"
-    ``````
-??? abstract "run_single_prompt.sh"
-    ``````sh
-    --8<-- "examples/offline_inference/qwen3_omni/run_single_prompt.sh"
-    ``````
-??? abstract "run_single_prompt_tp.sh"
-    ``````sh
-    --8<-- "examples/offline_inference/qwen3_omni/run_single_prompt_tp.sh"
-    ``````

docs/user_guide/examples/offline_inference/qwen_2_5_omni.md

@@ -0,0 +1,97 @@
+# Offline Example of vLLM-omni for Qwen2.5-omni
+
+Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/offline_inference/qwen_2_5_omni>.
+
+
+## 🛠️ Installation
+
+### Environment setup
+
+Use Docker to keep consistent basic environment (Optional, Recommended)
+```bash
+docker run --gpus all --ipc=host --network=host -v $source_dir:$container_dir --name $container_name -it nvcr.io/nvidia/pytorch:25.01-py3 bash
+```
+
+Set up basic uv environment
+```bash
+pip install uv
+uv venv --python 3.12 --seed
+source .venv/bin/activate
+```
+
+### Installation of vLLM (for users)
+
+Now we build it based on vLLM v0.11.0. Please install it with command below.
+```bash
+uv pip install vllm==0.11.0 --torch-backend=auto
+```
+
+### Installation of vLLM (for developers)
+
+If you want to check or debug with source code of vLLM, install stable release version of vllm with 0.11.0 from source with pre-built wheel file.
+
+```bash
+git clone https://github.com/vllm-project/vllm.git
+cd vllm
+git checkout v0.11.0
+```
+Set up environment variables to get pre-built wheels. If there are internet problems, just download the whl file manually. And set VLLM_PRECOMPILED_WHEEL_LOCATION as your local absolute path of whl file.
+```bash
+export VLLM_PRECOMPILED_WHEEL_LOCATION=https://github.com/vllm-project/vllm/releases/download/v0.11.0/vllm-0.11.0-cp38-abi3-manylinux1_x86_64.whl
+```
+Install vllm with command below (If you have no existing PyTorch).
+```bash
+uv pip install --editable .
+```
+
+Install vllm with command below (If you already have PyTorch).
+```bash
+python use_existing_torch.py
+uv pip install -r requirements/build.txt
+uv pip install --no-build-isolation --editable .
+```
+
+### Verification for successful installation of vLLM
+Just run the command below. If no error, it demonstrates that the installation is successful.
+```bash
+python -c "import vllm._C"
+```
+
+### Installation of vLLM-omni
+Install additional requirements for vllm-omni
+```bash
+git clone https://github.com/vllm-project/vllm-omni.git
+cd vllm_omni
+uv pip install -e .
+```
+
+## Run examples (Qwen2.5-omni)
+
+Get into the example folder
+```bash
+cd examples/offline_inference/qwen_2_5_omni
+```
+Then run
+```bash
+bash run.sh
+```
+The output audio is saved in ./output_audio
+
+## Example materials
+
+??? abstract "end2end.py"
+    ``````py
+    --8<-- "examples/offline_inference/qwen_2_5_omni/end2end.py"
+    ``````
+??? abstract "processing_omni.py"
+    ``````py
+    --8<-- "examples/offline_inference/qwen_2_5_omni/processing_omni.py"
+    ``````
+??? abstract "run.sh"
+    ``````sh
+    --8<-- "examples/offline_inference/qwen_2_5_omni/run.sh"
+    ``````
+??? abstract "utils.py"
+    ``````py
+    --8<-- "examples/offline_inference/qwen_2_5_omni/utils.py"
+    ``````

docs/user_guide/examples/offline_inference/text_to_image.md

@@ -1,56 +0,0 @@
-# Qwen-Image Offline Inference
-
-Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/offline_inference/text_to_image>.
-
-
-This folder provides two simple entrypoints for experimenting with `Qwen/Qwen-Image` `Tongyi-MAI/Z-Image-Turbo` using vLLM-Omni:
-
-- `text_to_image.py`: command-line script for single image generation.
-- `web_demo.py`: lightweight Gradio UI for interactive prompt/seed/CFG exploration.
-
-
-## Local CLI Usage
-
-```bash
-python text_to_image.py \
-  --model Tongyi-MAI/Z-Image-Turbo \
-  --prompt "a cup of coffee on the table" \
-  --seed 42 \
-  --cfg_scale 4.0 \
-  --num_images_per_prompt 1 \
-  --num_inference_steps 50 \
-  --height 1024 \
-  --width 1024 \
-  --output outputs/coffee.png
-```
-
-Key arguments:
-
-- `--prompt`: text description (string).
-- `--seed`: integer seed for deterministic sampling.
-- `--cfg_scale`: true CFG scale (model-specific guidance strength).
-- `--num_images_per_prompt`: number of images to generate per prompt (saves as `output`, `output_1`, ...).
-- `--num_inference_steps`: diffusion sampling steps (more steps = higher quality, slower).
-- `--height/--width`: output resolution (defaults 1024x1024).
-- `--output`: path to save the generated PNG.
-
-## Web UI Demo
-
-Launch the gradio demo:
-
-```bash
-python gradio_demo.py --port 7862
-```
-
-Then open `http://localhost:7862/` on your local browser to interact with the web UI.
-
-## Example materials
-
-??? abstract "gradio_demo.py"
-    ``````py
-    --8<-- "examples/offline_inference/text_to_image/gradio_demo.py"
-    ``````
-??? abstract "text_to_image.py"
-    ``````py
-    --8<-- "examples/offline_inference/text_to_image/text_to_image.py"
-    ``````

docs/user_guide/examples/offline_inference/text_to_video.md

@@ -1,38 +0,0 @@
-# Wan2.2 Text-to-Video Offline Inference
-
-Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/offline_inference/text_to_video>. Added in PR [#202](https://github.com/vllm-project/vllm-omni/pull/202).
-
-The `Wan-AI/Wan2.2-T2V-A14B-Diffusers` pipeline generates short videos from text prompts.
-
-## Local CLI Usage
-
-```bash
-python text_to_video.py \
-  --prompt "Two anthropomorphic cats in comfy boxing gear and bright gloves fight intensely on a spotlighted stage." \
-  --negative_prompt "<optional quality filter>" \
-  --height 480 \
-  --width 640 \
-  --num_frames 32 \
-  --guidance_scale 4.0 \
-  --guidance_scale_high 3.0 \
-  --num_inference_steps 40 \
-  --fps 16 \
-  --output t2v_out.mp4
-```
-
-Key arguments:
-- `--prompt`: text description (string).
-- `--height/--width`: output resolution (defaults 720x1280). Dimensions should align with Wan VAE downsampling (multiples of 8).
-- `--num_frames`: Number of frames (Wan default is 81).
-- `--guidance_scale` and `--guidance_scale_high`: CFG scale (applied to low/high)..
-- `--negative_prompt`: optional list of artifacts to suppress (the PR demo used a long Chinese string).
-- `--boundary_ratio`: Boundary split ratio for low/high DiT.
-- `--fps`: frames per second for the saved MP4 (requires `diffusers` export_to_video).
-- `--output`: path to save the generated video.
-
-## Example materials
-
-??? abstract "text_to_video.py"
-    ``````py
-    --8<-- "examples/offline_inference/wan22/text_to_video.py"
-    ``````

docs/user_guide/examples/online_serving/README.md

@@ -0,0 +1,35 @@
+# Online serving Example of vLLM-omni for Qwen2.5-omni
+
+Source <https://github.com/vllm-project/vllm-omni/blob/main/examples\online_serving\README.md>.
+
+
+## 🛠️ Installation
+
+Please refer to [README.md](https://github.com/vllm-project/vllm-omni/blob/main/README.md)
+
+## Run examples (Qwen2.5-omni)
+
+Launch the server
+```bash
+vllm serve Qwen/Qwen2.5-Omni-7B --omni --port 8091
+```
+
+If you have custom stage configs file, launch the server with command below
+```bash
+vllm serve Qwen/Qwen2.5-Omni-7B --omni --port 8091 --stage-configs-path /path/to/stage_configs_file
+```
+
+Get into the example folder
+```bash
+cd examples/online_serving
+```
+
+Send request via python
+```bash
+python openai_chat_completion_client_for_multimodal_generation.py
+```
+
+Send request via curl
+```bash
+bash run_curl_multimodal_generation.sh
+```

docs/user_guide/examples/online_serving/gradio_demo.md

@@ -1,7 +0,0 @@
-# Gradio Demo
-
-Source <https://github.com/vllm-project/vllm-omni/blob/main/examples/online_serving/gradio_demo.py>.
-
-``````py
---8<-- "examples/online_serving/gradio_demo.py"
-``````

docs/user_guide/examples/online_serving/index.md

@@ -0,0 +1,5 @@
+# Online Serving
+
+## Available Examples
+
+Online serving is under active development. Coming soon.

docs/user_guide/examples/online_serving/openai_chat_completion_client_for_multimodal_generation.md

@@ -1,7 +1,7 @@
 # OpenAI Chat Completion Client For Multimodal Generation
 
-Source <https://github.com/vllm-project/vllm-omni/blob/main/examples/online_serving/openai_chat_completion_client_for_multimodal_generation.py>.
+Source <https://github.com/vllm-project/vllm-omni/blob/main/examples\online_serving\openai_chat_completion_client_for_multimodal_generation.py>.
 
 ``````py
---8<-- "examples/online_serving/openai_chat_completion_client_for_multimodal_generation.py"
+--8<-- "examples\online_serving\openai_chat_completion_client_for_multimodal_generation.py"
 ``````

docs/user_guide/examples/online_serving/qwen2_5_omni.md

@@ -1,119 +0,0 @@
-# Online serving Example of vLLM-Omni for Qwen2.5-Omni
-
-Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/online_serving/qwen2_5_omni>.
-
-
-## 🛠️ Installation
-
-Please refer to [README.md](https://github.com/vllm-project/vllm-omni/tree/main/README.md)
-
-## Run examples (Qwen2.5-Omni)
-
-### Launch the Server
-
-```bash
-vllm serve Qwen/Qwen2.5-Omni-7B --omni --port 8091
-```
-
-If you have custom stage configs file, launch the server with command below
-```bash
-vllm serve Qwen/Qwen2.5-Omni-7B --omni --port 8091 --stage-configs-path /path/to/stage_configs_file
-```
-
-### Send Multi-modal Request
-
-Get into the example folder
-```bash
-cd examples/online_serving/qwen2_5_omni
-```
-
-####  Send request via python
-
-```bash
-python openai_chat_completion_client_for_multimodal_generation.py --query-type mixed_modalities
-```
-
-The Python client supports the following command-line arguments:
-
-- `--query-type` (or `-q`): Query type (default: `mixed_modalities`)
-  - Options: `mixed_modalities`, `use_audio_in_video`, `multi_audios`, `text`
-- `--video-path` (or `-v`): Path to local video file or URL
-  - If not provided and query-type uses video, uses default video URL
-  - Supports local file paths (automatically encoded to base64) or HTTP/HTTPS URLs
-  - Example: `--video-path /path/to/video.mp4` or `--video-path https://example.com/video.mp4`
-- `--image-path` (or `-i`): Path to local image file or URL
-  - If not provided and query-type uses image, uses default image URL
-  - Supports local file paths (automatically encoded to base64) or HTTP/HTTPS URLs
-  - Supports common image formats: JPEG, PNG, GIF, WebP
-  - Example: `--image-path /path/to/image.jpg` or `--image-path https://example.com/image.png`
-- `--audio-path` (or `-a`): Path to local audio file or URL
-  - If not provided and query-type uses audio, uses default audio URL
-  - Supports local file paths (automatically encoded to base64) or HTTP/HTTPS URLs
-  - Supports common audio formats: MP3, WAV, OGG, FLAC, M4A
-  - Example: `--audio-path /path/to/audio.wav` or `--audio-path https://example.com/audio.mp3`
-- `--prompt` (or `-p`): Custom text prompt/question
-  - If not provided, uses default prompt for the selected query type
-  - Example: `--prompt "What are the main activities shown in this video?"`
-
-
-For example, to use mixed modalities with all local files:
-
-```bash
-python openai_chat_completion_client_for_multimodal_generation.py \
-    --query-type mixed_modalities \
-    --video-path /path/to/your/video.mp4 \
-    --image-path /path/to/your/image.jpg \
-    --audio-path /path/to/your/audio.wav \
-    --prompt "Analyze all the media content and provide a comprehensive summary."
-```
-
-####  Send request via curl
-
-```bash
-bash run_curl_multimodal_generation.sh mixed_modalities
-```
-
-## Run Local Web UI Demo
-
-This Web UI demo allows users to interact with the model through a web browser.
-
-### Running Gradio Demo
-
-Once vllm and vllm-omni are installed, you can launch the web service built on AsyncOmni by
-
-```bash
-python gradio_demo.py  --model Qwen/Qwen2.5-Omni-7B --port 7861
-```
-
-Then open `http://localhost:7861/` on your local browser to interact with the web UI.
-
-The gradio script supports the following arguments:
-
-- `--model`: Model name
-- `--ip`: Host/IP for Gradio server (default: 127.0.0.1)
-- `--port`: Port for Gradio server (default: 7861)
-- `--stage-configs-path`: Path to custom stage configs YAML file (optional)
-- `--share`: Share the Gradio demo publicly (creates a public link)
-
-### FAQ
-
-If you encounter error about backend of librosa, try to install ffmpeg with command below.
-```
-sudo apt update
-sudo apt install ffmpeg
-```
-
-## Example materials
-
-??? abstract "gradio_demo.py"
-    ``````py
-    --8<-- "examples/online_serving/qwen2_5_omni/gradio_demo.py"
-    ``````
-??? abstract "openai_chat_completion_client_for_multimodal_generation.py"
-    ``````py
-    --8<-- "examples/online_serving/qwen2_5_omni/openai_chat_completion_client_for_multimodal_generation.py"
-    ``````
-??? abstract "run_curl_multimodal_generation.sh"
-    ``````sh
-    --8<-- "examples/online_serving/qwen2_5_omni/run_curl_multimodal_generation.sh"
-    ``````

docs/user_guide/examples/online_serving/qwen3_omni.md

@@ -1,124 +0,0 @@
-# Online serving Example of vLLM-Omni for Qwen3-omni
-
-Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/online_serving/qwen3_omni>.
-
-
-## 🛠️ Installation
-
-Please refer to [README.md](https://github.com/vllm-project/vllm-omni/tree/main/README.md)
-
-## Run examples (Qwen3-Omni)
-
-### Launch the Server
-
-```bash
-vllm serve Qwen/Qwen3-Omni-30B-A3B-Instruct --omni --port 8091
-```
-
-If you have custom stage configs file, launch the server with command below
-```bash
-vllm serve Qwen/Qwen3-Omni-30B-A3B-Instruct --omni --port 8091 --stage-configs-path /path/to/stage_configs_file
-```
-
-### Send Multi-modal Request
-
-Get into the example folder
-```bash
-cd examples/online_serving/qwen3_omni
-```
-
-####  Send request via python
-
-```bash
-python openai_chat_completion_client_for_multimodal_generation.py --query-type use_image
-```
-
-The Python client supports the following command-line arguments:
-
-- `--query-type` (or `-q`): Query type (default: `use_video`)
-  - Options: `text`, `use_audio`, `use_image`, `use_video`
-- `--model` (or `-m`): Model name/path (default: `Qwen/Qwen3-Omni-30B-A3B-Instruct`)
-- `--video-path` (or `-v`): Path to local video file or URL
-  - If not provided and query-type is `use_video`, uses default video URL
-  - Supports local file paths (automatically encoded to base64) or HTTP/HTTPS URLs
-  - Example: `--video-path /path/to/video.mp4` or `--video-path https://example.com/video.mp4`
-- `--image-path` (or `-i`): Path to local image file or URL
-  - If not provided and query-type is `use_image`, uses default image URL
-  - Supports local file paths (automatically encoded to base64) or HTTP/HTTPS URLs
-  - Supports common image formats: JPEG, PNG, GIF, WebP
-  - Example: `--image-path /path/to/image.jpg` or `--image-path https://example.com/image.png`
-- `--audio-path` (or `-a`): Path to local audio file or URL
-  - If not provided and query-type is `use_audio`, uses default audio URL
-  - Supports local file paths (automatically encoded to base64) or HTTP/HTTPS URLs
-  - Supports common audio formats: MP3, WAV, OGG, FLAC, M4A
-  - Example: `--audio-path /path/to/audio.wav` or `--audio-path https://example.com/audio.mp3`
-- `--prompt` (or `-p`): Custom text prompt/question
-  - If not provided, uses default prompt for the selected query type
-  - Example: `--prompt "What are the main activities shown in this video?"`
-
-
-For example, to use a local video file with custom prompt:
-
-```bash
-python openai_chat_completion_client_for_multimodal_generation.py \
-    --query-type use_video \
-    --video-path /path/to/your/video.mp4 \
-    --prompt "What are the main activities shown in this video?"
-```
-
-####  Send request via curl
-
-```bash
-bash run_curl_multimodal_generation.sh use_image
-```
-
-
-### FAQ
-
-If you encounter error about backend of librosa, try to install ffmpeg with command below.
-```
-sudo apt update
-sudo apt install ffmpeg
-```
-
-## Run Local Web UI Demo
-
-This Web UI demo allows users to interact with the model through a web browser.
-
-### Running Gradio Demo
-
-Once vllm and vllm-omni are installed, you can launch the web service built on AsyncOmni by
-
-```bash
-python gradio_demo.py  --model Qwen/Qwen3-Omni-30B-A3B-Instruct --port 7861
-```
-
-Then open `http://localhost:7861/` on your local browser to interact with the web UI.
-
-
-The gradio script supports the following arguments:
-
-- `--model`: Model name
-- `--ip`: Host/IP for Gradio server (default: 127.0.0.1)
-- `--port`: Port for Gradio server (default: 7861)
-- `--stage-configs-path`: Path to custom stage configs YAML file (optional)
-- `--share`: Share the Gradio demo publicly (creates a public link)
-
-## Example materials
-
-??? abstract "gradio_demo.py"
-    ``````py
-    --8<-- "examples/online_serving/qwen3_omni/gradio_demo.py"
-    ``````
-??? abstract "openai_chat_completion_client_for_multimodal_generation.py"
-    ``````py
-    --8<-- "examples/online_serving/qwen3_omni/openai_chat_completion_client_for_multimodal_generation.py"
-    ``````
-??? abstract "qwen3_omni_moe_thinking.yaml"
-    ``````yaml
-    --8<-- "examples/online_serving/qwen3_omni/qwen3_omni_moe_thinking.yaml"
-    ``````
-??? abstract "run_curl_multimodal_generation.sh"
-    ``````sh
-    --8<-- "examples/online_serving/qwen3_omni/run_curl_multimodal_generation.sh"
-    ``````

docs/user_guide/getting_started/quickstart.md

@@ -0,0 +1,55 @@
+# Quickstart
+
+This guide will help you get started with vLLM-omni.
+
+## Installation
+
+Use Docker to keep consistent basic environment (Optional, Recommended)
+
+```bash
+docker run --gpus all --ipc=host --network=host -v $source_dir:$container_dir --rm --name $container_name -it nvcr.io/nvidia/pytorch:25.01-py3 bash
+```
+
+Set up basic uv environment
+
+```bash
+pip install uv
+uv venv --python 3.12 --seed
+source .venv/bin/activate
+```
+
+Install vLLM
+
+```bash
+uv pip install vllm==0.11.0 --torch-backend=auto
+```
+
+Install vLLM-omni
+
+```bash
+git clone https://github.com/vllm-project/vllm-omni.git
+cd vllm_omni
+uv pip install -e .
+```
+
+## Run examples (Qwen2.5-omni)
+
+Get into the example folder
+
+```bash
+cd examples/offline_inference/qwen_2_5_omni
+```
+
+Then run.
+
+```bash
+bash run.sh
+```
+
+The output audio is saved in ./output_audio
+
+## Next Steps
+
+- Read the [architecture documentation](../../contributing/design_documents/vllm_omni_design.md)
+- Check out the [API reference](../../api/overview.md)
+- Explore the [examples](../examples/index.md)

docs/user_guide/teacache.md

@@ -1,99 +0,0 @@
-# TeaCache Configuration Guide
-
-TeaCache speeds up diffusion model inference by caching transformer computations when consecutive timesteps are similar. This typically provides **1.5x-2.0x speedup** with minimal quality loss.
-
-## Quick Start
-
-Enable TeaCache by setting `cache_adapter` to `"tea_cache"`:
-
-```python
-from vllm_omni import Omni
-
-# Simple configuration - model_class_name and model_type are auto-detected
-omni = Omni(
-    model="Qwen/Qwen-Image",
-    cache_adapter="tea_cache",
-    cache_config={
-        "rel_l1_thresh": 0.2  # Optional, defaults to 0.2
-    }
-)
-outputs = omni.generate(prompt="A cat sitting on a windowsill", num_inference_steps=50)
-```
-
-### Using Environment Variable
-
-You can also enable TeaCache via environment variable:
-
-```bash
-export DIFFUSION_CACHE_ADAPTER=tea_cache
-```
-
-Then initialize without explicitly setting `cache_adapter`:
-
-```python
-from vllm_omni import Omni
-
-omni = Omni(
-    model="Qwen/Qwen-Image",
-    cache_config={"rel_l1_thresh": 0.2}  # Optional
-)
-```
-
-## Configuration Parameters
-
-### `rel_l1_thresh` (float, default: `0.2`)
-
-Controls the balance between speed and quality. Lower values prioritize quality, higher values prioritize speed.
-
-**Recommended values:**
-
-- `0.2` - **~1.5x speedup** with minimal quality loss (recommended)
-- `0.4` - **~1.8x speedup** with slight quality loss
-- `0.6` - **~2.0x speedup** with noticeable quality loss
-- `0.8` - **~2.25x speedup** with significant quality loss
-
-## Examples
-
-### Python API
-
-```python
-from vllm_omni import Omni
-
-omni = Omni(
-    model="Qwen/Qwen-Image",
-    cache_adapter="tea_cache",
-    cache_config={"rel_l1_thresh": 0.2}
-)
-outputs = omni.generate(prompt="A cat sitting on a windowsill", num_inference_steps=50)
-```
-
-## Performance Tuning
-
-Start with the default `rel_l1_thresh=0.2` and adjust based on your needs:
-
-- **Maximum quality**: Use `0.1-0.2`
-- **Balanced**: Use `0.2-0.4` (recommended)
-- **Maximum speed**: Use `0.6-0.8` (may reduce quality)
-
-## Troubleshooting
-
-### Quality Degradation
-
-If you notice quality issues, lower the threshold:
-
-```python
-cache_config={"rel_l1_thresh": 0.1}  # More conservative caching
-```
-
-## Supported Models
-
-Currently supported models:
-
-- **QwenImagePipeline**
-  - Model identifier: `"Qwen/Qwen-Image"`
-  - Example: `model="Qwen/Qwen-Image"`
-
-Future support (extractors in development):
-
-- **FluxPipeline** - Coming soon
-- **CogVideoXPipeline** - Coming soon

examples/offline_inference/image_to_image/image_edit.py

@@ -1,143 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-
-"""
-Example script for image editing with Qwen-Image-Edit.
-
-Usage:
-    python image_edit.py \
-        --image input.png \
-        --prompt "Let this mascot dance under the moon, surrounded by floating stars and poetic bubbles such as 'Be Kind'" \
-        --output output_image_edit.png \
-        --num_inference_steps 50 \
-        --cfg_scale 4.0
-
-For more options, run:
-    python image_edit.py --help
-"""
-
-import argparse
-import os
-from pathlib import Path
-
-import torch
-from PIL import Image
-
-from vllm_omni.entrypoints.omni import Omni
-from vllm_omni.utils.platform_utils import detect_device_type, is_npu
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="Edit an image with Qwen-Image-Edit.")
-    parser.add_argument(
-        "--model",
-        default="Qwen/Qwen-Image-Edit",
-        help="Diffusion model name or local path.",
-    )
-    parser.add_argument(
-        "--image",
-        type=str,
-        required=True,
-        help="Path to input image file (PNG, JPG, etc.).",
-    )
-    parser.add_argument(
-        "--prompt",
-        type=str,
-        required=True,
-        help="Text prompt describing the edit to make to the image.",
-    )
-    parser.add_argument(
-        "--negative_prompt",
-        type=str,
-        default=" ",
-        required=False,
-    )
-    parser.add_argument(
-        "--seed",
-        type=int,
-        default=0,
-        help="Random seed for deterministic results.",
-    )
-    parser.add_argument(
-        "--cfg_scale",
-        type=float,
-        default=4.0,
-        help="True classifier-free guidance scale specific to Qwen-Image-Edit.",
-    )
-    parser.add_argument(
-        "--output",
-        type=str,
-        default="output_image_edit.png",
-        help="Path to save the edited image (PNG).",
-    )
-    parser.add_argument(
-        "--num_outputs_per_prompt",
-        type=int,
-        default=1,
-        help="Number of images to generate for the given prompt.",
-    )
-    parser.add_argument(
-        "--num_inference_steps",
-        type=int,
-        default=50,
-        help="Number of denoising steps for the diffusion sampler.",
-    )
-    return parser.parse_args()
-
-
-def main():
-    args = parse_args()
-
-    # Validate input image exists
-    if not os.path.exists(args.image):
-        raise FileNotFoundError(f"Input image not found: {args.image}")
-
-    # Load input image
-    input_image = Image.open(args.image).convert("RGB")
-    print(f"Loaded input image from {args.image} (size: {input_image.size})")
-
-    device = detect_device_type()
-    generator = torch.Generator(device=device).manual_seed(args.seed)
-
-    # Enable VAE memory optimizations on NPU
-    vae_use_slicing = is_npu()
-    vae_use_tiling = is_npu()
-
-    # Initialize Omni with QwenImageEditPipeline
-    omni = Omni(
-        model=args.model,
-        model_class_name="QwenImageEditPipeline",
-        vae_use_slicing=vae_use_slicing,
-        vae_use_tiling=vae_use_tiling,
-    )
-    print("Pipeline loaded")
-
-    # Generate edited image
-    images = omni.generate(
-        prompt=args.prompt,
-        pil_image=input_image,
-        negative_prompt=args.negative_prompt,
-        generator=generator,
-        true_cfg_scale=args.cfg_scale,
-        num_inference_steps=args.num_inference_steps,
-        num_outputs_per_prompt=args.num_outputs_per_prompt,
-    )
-
-    # Save output image(s)
-    output_path = Path(args.output)
-    output_path.parent.mkdir(parents=True, exist_ok=True)
-    suffix = output_path.suffix or ".png"
-    stem = output_path.stem or "output_image_edit"
-
-    if args.num_outputs_per_prompt <= 1:
-        images[0].save(output_path)
-        print(f"Saved edited image to {os.path.abspath(output_path)}")
-    else:
-        for idx, img in enumerate(images):
-            save_path = output_path.parent / f"{stem}_{idx}{suffix}"
-            img.save(save_path)
-            print(f"Saved edited image to {os.path.abspath(save_path)}")
-
-
-if __name__ == "__main__":
-    main()

examples/offline_inference/image_to_image/image_to_image.md

@@ -1,30 +0,0 @@
-# Qwen-Image-Edit Offline Inference
-
-Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/offline_inference/image_to_image>. Added in PR [#196](https://github.com/vllm-project/vllm-omni/pull/196).
-
-This example edits an input image with `Qwen/Qwen-Image-Edit` using the `image_edit.py` CLI.
-
-## Local CLI Usage
-
-```bash
-python image_edit.py \
-  --image qwen_bear.png \
-  --prompt "Let this mascot dance under the moon, surrounded by floating stars and poetic bubbles such as 'Be Kind'" \
-  --output output_image_edit.png \
-  --num_inference_steps 50 \
-  --cfg_scale 4.0
-```
-
-Key arguments:
-- `--image`: path to the source image (PNG/JPG, converted to RGB).
-- `--prompt` / `--negative_prompt`: text description (string).
-- `--cfg_scale`: true CFG scale for Qwen-Image-Edit (quality vs. fidelity).
-- `--num_inference_steps`: diffusion sampling steps (more steps = higher quality, slower).
-- `--output`: path to save the generated PNG.
-
-## Example materials
-
-??? abstract "image_edit.py"
-    ``````py
-    --8<-- "examples/offline_inference/image_to_image/image_edit.py"
-    ``````

examples/offline_inference/qwen2_5_omni/README.md

@@ -1,11 +1,10 @@
-# Offline Example of vLLM-Omni for Qwen2.5-Omni
+# Offline Example of vLLM-omni for Qwen2.5-omni
 
 ## 🛠️ Installation
 
 Please refer to [README.md](../../../README.md)
 
-## Run examples (Qwen2.5-Omni)
-
+## Run examples (Qwen2.5-omni)
 ### Multiple Prompts
 Download dataset from [seed_tts](https://drive.google.com/file/d/1GlSjVfSHkW3-leKKBlfrjuuTGqQ_xaLP/edit). To get the prompt, you can:
 ```bash
@@ -24,8 +23,7 @@ Then run the command below.
 ```bash
 bash run_multiple_prompts.sh
 ```
-
-### Single Prompt
+### Single Prompts
 Get into the example folder
 ```bash
 cd examples/offline_inference/qwen2_5_omni
@@ -34,39 +32,3 @@ Then run the command below.
 ```bash
 bash run_single_prompt.sh
 ```
-
-#### Using Local Media Files
-The `end2end.py` script supports local media files (audio, video, image) via CLI arguments:
-
-```bash
-# Use single local media files
-python end2end.py --query-type use_image --image-path /path/to/image.jpg
-python end2end.py --query-type use_video --video-path /path/to/video.mp4
-python end2end.py --query-type use_audio --audio-path /path/to/audio.wav
-
-# Combine multiple local media files
-python end2end.py --query-type mixed_modalities \
-    --video-path /path/to/video.mp4 \
-    --image-path /path/to/image.jpg \
-    --audio-path /path/to/audio.wav
-
-# Use audio from video file
-python end2end.py --query-type use_audio_in_video --video-path /path/to/video.mp4
-
-```
-
-If media file paths are not provided, the script will use default assets. Supported query types:
-- `use_image`: Image input only
-- `use_video`: Video input only
-- `use_audio`: Audio input only
-- `mixed_modalities`: Audio + image + video
-- `use_audio_in_video`: Extract audio from video
-- `text`: Text-only query
-
-### FAQ
-
-If you encounter error about backend of librosa, try to install ffmpeg with command below.
-```
-sudo apt update
-sudo apt install ffmpeg
-```

examples/offline_inference/qwen2_5_omni/end2end.py

@@ -1,326 +1,167 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""
-This example shows how to use vLLM-Omni for running offline inference
-with the correct prompt format on Qwen2.5-Omni
-"""
-
+import argparse
 import os
-from typing import NamedTuple, Optional
+import os as _os_env_toggle
+import random
 
-import librosa
 import numpy as np
 import soundfile as sf
-from PIL import Image
-from vllm.assets.audio import AudioAsset
-from vllm.assets.image import ImageAsset
-from vllm.assets.video import VideoAsset, video_to_ndarrays
-from vllm.multimodal.image import convert_image_mode
+import torch
+from utils import make_omni_prompt
 from vllm.sampling_params import SamplingParams
-from vllm.utils import FlexibleArgumentParser
 
-from vllm_omni.entrypoints.omni import Omni
+from vllm_omni.entrypoints.omni_llm import OmniLLM
+
+_os_env_toggle.environ["VLLM_USE_V1"] = "1"
 
 SEED = 42
+# Set all random seeds
+random.seed(SEED)
+np.random.seed(SEED)
+torch.manual_seed(SEED)
+torch.cuda.manual_seed(SEED)
+torch.cuda.manual_seed_all(SEED)
+
+# Make PyTorch deterministic
+torch.backends.cudnn.deterministic = True
+torch.backends.cudnn.benchmark = False
+
+# Set environment variables for deterministic behavior
+os.environ["PYTHONHASHSEED"] = str(SEED)
+os.environ["CUBLAS_WORKSPACE_CONFIG"] = ":4096:8"
 
 
-class QueryResult(NamedTuple):
-    inputs: dict
-    limit_mm_per_prompt: dict[str, int]
-
-
-# NOTE: The default `max_num_seqs` and `max_model_len` may result in OOM on
-# lower-end GPUs.
-# Unless specified, these settings have been tested to work on a single L4.
-
-default_system = (
-    "You are Qwen, a virtual human developed by the Qwen Team, Alibaba "
-    "Group, capable of perceiving auditory and visual inputs, as well as "
-    "generating text and speech."
-)
-
-
-def get_text_query(question: str = None) -> QueryResult:
-    if question is None:
-        question = "Explain the system architecture for a scalable audio generation pipeline. Answer in 15 words."
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
+def parse_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--model",
+        required=True,
+        help="Path to merged model directory (will be created if downloading).",
     )
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-        },
-        limit_mm_per_prompt={},
+    parser.add_argument("--thinker-model", type=str, default=None)
+    parser.add_argument("--talker-model", type=str, default=None)
+    parser.add_argument("--code2wav-model", type=str, default=None)
+    parser.add_argument(
+        "--hf-hub-id",
+        default="Qwen/Qwen2.5-Omni-7B",
+        help="Hugging Face repo id to download if needed.",
+    )
+    parser.add_argument("--hf-revision", default=None, help="Optional HF revision (branch/tag/commit).")
+    parser.add_argument("--prompts", nargs="+", default=None, help="Input text prompts.")
+    parser.add_argument("--voice-type", default="default", help="Voice type, e.g., m02, f030, default.")
+    parser.add_argument(
+        "--code2wav-dir",
+        default=None,
+        help="Path to code2wav folder (contains spk_dict.pt).",
+    )
+    parser.add_argument("--dit-ckpt", default=None, help="Path to DiT checkpoint file (e.g., dit.pt).")
+    parser.add_argument("--bigvgan-ckpt", default=None, help="Path to BigVGAN checkpoint file.")
+    parser.add_argument("--dtype", default="bfloat16", choices=["float16", "bfloat16", "float32"])
+    parser.add_argument("--max-model-len", type=int, default=32768)
+    parser.add_argument(
+        "--init-sleep-seconds",
+        type=int,
+        default=20,
+        help="Sleep seconds after starting each stage process to allow initialization (default: 20)",
     )
 
-
-def get_mixed_modalities_query(
-    video_path: Optional[str] = None,
-    image_path: Optional[str] = None,
-    audio_path: Optional[str] = None,
-    num_frames: int = 16,
-    sampling_rate: int = 16000,
-) -> QueryResult:
-    question = "What is recited in the audio? What is the content of this image? Why is this video funny?"
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|audio_bos|><|AUDIO|><|audio_eos|>"
-        "<|vision_bos|><|IMAGE|><|vision_eos|>"
-        "<|vision_bos|><|VIDEO|><|vision_eos|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
+    parser.add_argument("--thinker-only", action="store_true")
+    parser.add_argument("--text-only", action="store_true")
+    parser.add_argument("--do-wave", action="store_true")
+    parser.add_argument(
+        "--prompt_type",
+        choices=[
+            "text",
+            "audio",
+            "audio-long",
+            "audio-long-chunks",
+            "audio-long-expand-chunks",
+            "image",
+            "video",
+            "video-frames",
+            "audio-in-video",
+            "audio-in-video-v2",
+            "audio-multi-round",
+            "badcase-vl",
+            "badcase-text",
+            "badcase-image-early-stop",
+            "badcase-two-audios",
+            "badcase-two-videos",
+            "badcase-multi-round",
+            "badcase-voice-type",
+            "badcase-voice-type-v2",
+            "badcase-audio-tower-1",
+            "badcase-audio-only",
+        ],
+        default="text",
     )
-
-    # Load video
-    if video_path:
-        if not os.path.exists(video_path):
-            raise FileNotFoundError(f"Video file not found: {video_path}")
-        video_frames = video_to_ndarrays(video_path, num_frames=num_frames)
-    else:
-        video_frames = VideoAsset(name="baby_reading", num_frames=num_frames).np_ndarrays
-
-    # Load image
-    if image_path:
-        if not os.path.exists(image_path):
-            raise FileNotFoundError(f"Image file not found: {image_path}")
-        pil_image = Image.open(image_path)
-        image_data = convert_image_mode(pil_image, "RGB")
-    else:
-        image_data = convert_image_mode(ImageAsset("cherry_blossom").pil_image, "RGB")
-
-    # Load audio
-    if audio_path:
-        if not os.path.exists(audio_path):
-            raise FileNotFoundError(f"Audio file not found: {audio_path}")
-        audio_signal, sr = librosa.load(audio_path, sr=sampling_rate)
-        audio_data = (audio_signal.astype(np.float32), sr)
-    else:
-        audio_data = AudioAsset("mary_had_lamb").audio_and_sample_rate
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "audio": audio_data,
-                "image": image_data,
-                "video": video_frames,
-            },
-        },
-        limit_mm_per_prompt={"audio": 1, "image": 1, "video": 1},
+    parser.add_argument("--use-torchvision", action="store_true")
+    parser.add_argument("--tokenize", action="store_true")
+    parser.add_argument(
+        "--output-wav",
+        default="output.wav",
+        help="[Deprecated] Output wav directory (use --output-dir).",
     )
-
-
-def get_use_audio_in_video_query(
-    video_path: Optional[str] = None, num_frames: int = 16, sampling_rate: int = 16000
-) -> QueryResult:
-    question = "Describe the content of the video, then convert what the baby say into text."
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|vision_bos|><|VIDEO|><|vision_eos|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
+    parser.add_argument(
+        "--output-dir",
+        default="outputs",
+        help="Output directory to save text and wav files together.",
     )
-
-    if video_path:
-        if not os.path.exists(video_path):
-            raise FileNotFoundError(f"Video file not found: {video_path}")
-        video_frames = video_to_ndarrays(video_path, num_frames=num_frames)
-        # Extract audio from video file
-        audio_signal, sr = librosa.load(video_path, sr=sampling_rate)
-        audio = (audio_signal.astype(np.float32), sr)
-    else:
-        asset = VideoAsset(name="baby_reading", num_frames=num_frames)
-        video_frames = asset.np_ndarrays
-        audio = asset.get_audio(sampling_rate=sampling_rate)
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "video": video_frames,
-                "audio": audio,
-            },
-            "mm_processor_kwargs": {
-                "use_audio_in_video": True,
-            },
-        },
-        limit_mm_per_prompt={"audio": 1, "video": 1},
+    parser.add_argument(
+        "--thinker-hidden-states-dir",
+        default="thinker_hidden_states",
+        help="Path to thinker hidden states directory.",
     )
-
-
-def get_multi_audios_query(audio_path: Optional[str] = None, sampling_rate: int = 16000) -> QueryResult:
-    question = "Are these two audio clips the same?"
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|audio_bos|><|AUDIO|><|audio_eos|>"
-        "<|audio_bos|><|AUDIO|><|audio_eos|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
+    parser.add_argument(
+        "--batch-timeout",
+        type=int,
+        default=5,
+        help="Timeout for batching in seconds (default: 5)",
     )
-
-    if audio_path:
-        if not os.path.exists(audio_path):
-            raise FileNotFoundError(f"Audio file not found: {audio_path}")
-        audio_signal, sr = librosa.load(audio_path, sr=sampling_rate)
-        audio_data = (audio_signal.astype(np.float32), sr)
-        # Use the provided audio as the first audio, default as second
-        audio_list = [
-            audio_data,
-            AudioAsset("mary_had_lamb").audio_and_sample_rate,
-        ]
-    else:
-        audio_list = [
-            AudioAsset("winning_call").audio_and_sample_rate,
-            AudioAsset("mary_had_lamb").audio_and_sample_rate,
-        ]
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "audio": audio_list,
-            },
-        },
-        limit_mm_per_prompt={
-            "audio": 2,
-        },
+    parser.add_argument(
+        "--init-timeout",
+        type=int,
+        default=300,
+        help="Timeout for initializing stages in seconds (default: 300)",
     )
-
-
-def get_image_query(question: str = None, image_path: Optional[str] = None) -> QueryResult:
-    if question is None:
-        question = "What is the content of this image?"
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|vision_bos|><|IMAGE|><|vision_eos|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
+    parser.add_argument(
+        "--shm-threshold-bytes",
+        type=int,
+        default=65536,
+        help="Threshold for using shared memory in bytes (default: 65536)",
     )
-
-    if image_path:
-        if not os.path.exists(image_path):
-            raise FileNotFoundError(f"Image file not found: {image_path}")
-        pil_image = Image.open(image_path)
-        image_data = convert_image_mode(pil_image, "RGB")
-    else:
-        image_data = convert_image_mode(ImageAsset("cherry_blossom").pil_image, "RGB")
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "image": image_data,
-            },
-        },
-        limit_mm_per_prompt={"image": 1},
+    parser.add_argument(
+        "--enable-stats",
+        action="store_true",
+        default=False,
+        help="Enable writing detailed statistics (default: disabled)",
     )
-
-
-def get_video_query(question: str = None, video_path: Optional[str] = None, num_frames: int = 16) -> QueryResult:
-    if question is None:
-        question = "Why is this video funny?"
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|vision_bos|><|VIDEO|><|vision_eos|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
-    )
-
-    if video_path:
-        if not os.path.exists(video_path):
-            raise FileNotFoundError(f"Video file not found: {video_path}")
-        video_frames = video_to_ndarrays(video_path, num_frames=num_frames)
-    else:
-        video_frames = VideoAsset(name="baby_reading", num_frames=num_frames).np_ndarrays
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "video": video_frames,
-            },
-        },
-        limit_mm_per_prompt={"video": 1},
+    parser.add_argument(
+        "--txt-prompts",
+        type=str,
+        default=None,
+        help="Path to a .txt file with one prompt per line (preferred).",
     )
+    args = parser.parse_args()
+    return args
 
 
-def get_audio_query(question: str = None, audio_path: Optional[str] = None, sampling_rate: int = 16000) -> QueryResult:
-    if question is None:
-        question = "What is the content of this audio?"
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|audio_bos|><|AUDIO|><|audio_eos|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
-    )
+def main():
+    args = parse_args()
+    model_name = args.model
+    try:
+        # Preferred: load from txt file (one prompt per line)
+        if getattr(args, "txt_prompts", None) and args.prompt_type == "text":
+            with open(args.txt_prompts, encoding="utf-8") as f:
+                lines = [ln.strip() for ln in f.readlines()]
+            args.prompts = [ln for ln in lines if ln != ""]
+            print(f"[Info] Loaded {len(args.prompts)} prompts from {args.txt_prompts}")
+    except Exception as e:
+        print(f"[Error] Failed to load prompts: {e}")
+        raise
 
-    if audio_path:
-        if not os.path.exists(audio_path):
-            raise FileNotFoundError(f"Audio file not found: {audio_path}")
-        audio_signal, sr = librosa.load(audio_path, sr=sampling_rate)
-        audio_data = (audio_signal.astype(np.float32), sr)
-    else:
-        audio_data = AudioAsset("mary_had_lamb").audio_and_sample_rate
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "audio": audio_data,
-            },
-        },
-        limit_mm_per_prompt={"audio": 1},
-    )
-
-
-query_map = {
-    "mixed_modalities": get_mixed_modalities_query,
-    "use_audio_in_video": get_use_audio_in_video_query,
-    "multi_audios": get_multi_audios_query,
-    "use_image": get_image_query,
-    "use_video": get_video_query,
-    "use_audio": get_audio_query,
-    "text": get_text_query,
-}
-
-
-def main(args):
-    model_name = "Qwen/Qwen2.5-Omni-7B"
-
-    # Get paths from args
-    video_path = getattr(args, "video_path", None)
-    image_path = getattr(args, "image_path", None)
-    audio_path = getattr(args, "audio_path", None)
-    num_frames = getattr(args, "num_frames", 16)
-    sampling_rate = getattr(args, "sampling_rate", 16000)
-
-    # Get the query function and call it with appropriate parameters
-    query_func = query_map[args.query_type]
-    if args.query_type == "mixed_modalities":
-        query_result = query_func(
-            video_path=video_path,
-            image_path=image_path,
-            audio_path=audio_path,
-            num_frames=num_frames,
-            sampling_rate=sampling_rate,
-        )
-    elif args.query_type == "use_audio_in_video":
-        query_result = query_func(video_path=video_path, num_frames=num_frames, sampling_rate=sampling_rate)
-    elif args.query_type == "multi_audios":
-        query_result = query_func(audio_path=audio_path, sampling_rate=sampling_rate)
-    elif args.query_type == "use_image":
-        query_result = query_func(image_path=image_path)
-    elif args.query_type == "use_video":
-        query_result = query_func(video_path=video_path, num_frames=num_frames)
-    elif args.query_type == "use_audio":
-        query_result = query_func(audio_path=audio_path, sampling_rate=sampling_rate)
-    else:
-        query_result = query_func()
-
-    omni_llm = Omni(
+    if args.prompts is None:
+        raise ValueError("No prompts provided. Use --prompts ... or --txt-prompts <file.txt> (with --prompt_type text)")
+    omni_llm = OmniLLM(
         model=model_name,
         log_stats=args.enable_stats,
         log_file=("omni_llm_pipeline.log" if args.enable_stats else None),
@@ -364,16 +205,8 @@ def main(args):
         code2wav_sampling_params,
     ]
 
-    if args.txt_prompts is None:
-        prompts = [query_result.inputs for _ in range(args.num_prompts)]
-    else:
-        assert args.query_type == "text", "txt-prompts is only supported for text query type"
-        with open(args.txt_prompts, encoding="utf-8") as f:
-            lines = [ln.strip() for ln in f.readlines()]
-            prompts = [get_text_query(ln).inputs for ln in lines if ln != ""]
-            print(f"[Info] Loaded {len(prompts)} prompts from {args.txt_prompts}")
-
-    omni_outputs = omni_llm.generate(prompts, sampling_params_list)
+    prompt = [make_omni_prompt(args, prompt) for prompt in args.prompts]
+    omni_outputs = omni_llm.generate(prompt, sampling_params_list)
 
     # Determine output directory: prefer --output-dir; fallback to --output-wav
     output_dir = args.output_dir if getattr(args, "output_dir", None) else args.output_wav
@@ -384,7 +217,7 @@ def main(args):
                 request_id = int(output.request_id)
                 text_output = output.outputs[0].text
                 # Save aligned text file per request
-                prompt_text = prompts[request_id]["prompt"]
+                prompt_text = args.prompts[request_id]
                 out_txt = os.path.join(output_dir, f"{request_id:05d}.txt")
                 lines = []
                 lines.append("Prompt:\n")
@@ -406,108 +239,5 @@ def main(args):
                 print(f"Request ID: {request_id}, Saved audio to {output_wav}")
 
 
-def parse_args():
-    parser = FlexibleArgumentParser(description="Demo on using vLLM for offline inference with audio language models")
-    parser.add_argument(
-        "--query-type",
-        "-q",
-        type=str,
-        default="mixed_modalities",
-        choices=query_map.keys(),
-        help="Query type.",
-    )
-    parser.add_argument(
-        "--enable-stats",
-        action="store_true",
-        default=False,
-        help="Enable writing detailed statistics (default: disabled)",
-    )
-    parser.add_argument(
-        "--init-sleep-seconds",
-        type=int,
-        default=20,
-        help="Sleep seconds after starting each stage process to allow initialization (default: 20)",
-    )
-    parser.add_argument(
-        "--batch-timeout",
-        type=int,
-        default=5,
-        help="Timeout for batching in seconds (default: 5)",
-    )
-    parser.add_argument(
-        "--init-timeout",
-        type=int,
-        default=300,
-        help="Timeout for initializing stages in seconds (default: 300)",
-    )
-    parser.add_argument(
-        "--shm-threshold-bytes",
-        type=int,
-        default=65536,
-        help="Threshold for using shared memory in bytes (default: 65536)",
-    )
-    parser.add_argument(
-        "--output-wav",
-        default="output_audio",
-        help="[Deprecated] Output wav directory (use --output-dir).",
-    )
-    parser.add_argument(
-        "--num-prompts",
-        type=int,
-        default=1,
-        help="Number of prompts to generate.",
-    )
-    parser.add_argument(
-        "--txt-prompts",
-        type=str,
-        default=None,
-        help="Path to a .txt file with one prompt per line (preferred).",
-    )
-    parser.add_argument(
-        "--video-path",
-        "-v",
-        type=str,
-        default=None,
-        help="Path to local video file. If not provided, uses default video asset.",
-    )
-    parser.add_argument(
-        "--image-path",
-        "-i",
-        type=str,
-        default=None,
-        help="Path to local image file. If not provided, uses default image asset.",
-    )
-    parser.add_argument(
-        "--audio-path",
-        "-a",
-        type=str,
-        default=None,
-        help="Path to local audio file. If not provided, uses default audio asset.",
-    )
-    parser.add_argument(
-        "--num-frames",
-        type=int,
-        default=16,
-        help="Number of frames to extract from video (default: 16).",
-    )
-    parser.add_argument(
-        "--sampling-rate",
-        type=int,
-        default=16000,
-        help="Sampling rate for audio loading (default: 16000).",
-    )
-    parser.add_argument(
-        "--worker-backend", type=str, default="multi_process", choices=["multi_process", "ray"], help="backend"
-    )
-    parser.add_argument(
-        "--ray-address",
-        type=str,
-        default=None,
-        help="Address of the Ray cluster.",
-    )
-    return parser.parse_args()
-
-
 if __name__ == "__main__":
-    args = parse_args()
-    main(args)
+    main()

examples/offline_inference/qwen2_5_omni/extract_prompts.py

@@ -4,7 +4,7 @@ from typing import Optional
 
 
 def extract_prompt(line: str) -> Optional[str]:
-    # Extract the content between the first '|' and the second '|'
+    # 提取第一个 '|' 与第二个 '|' 之间的内容
     i = line.find("|")
     if i == -1:
         return None

examples/offline_inference/qwen2_5_omni/processing_omni.py

@@ -0,0 +1,367 @@
+from __future__ import annotations
+
+import base64
+import logging
+import math
+import os
+import time
+import warnings
+from functools import lru_cache
+from io import BytesIO
+
+import requests
+import torch
+import torchvision
+from packaging import version
+from PIL import Image
+from torchvision import io, transforms
+from torchvision.transforms import InterpolationMode
+
+logger = logging.getLogger(__name__)
+
+IMAGE_FACTOR = 28
+MIN_PIXELS = 4 * 28 * 28
+MAX_PIXELS = 16384 * 28 * 28
+MAX_RATIO = 200
+
+VIDEO_MIN_PIXELS = 128 * 28 * 28
+VIDEO_MAX_PIXELS = 768 * 28 * 28
+VIDEO_TOTAL_PIXELS = 24576 * 28 * 28
+FRAME_FACTOR = 2
+FPS = 2.0
+FPS_MIN_FRAMES = 4
+FPS_MAX_FRAMES = 768
+
+temporal_patch_size = 2
+spatial_patch_size = 14
+spatial_merge_size = 2
+
+
+def round_by_factor(number: int, factor: int) -> int:
+    """Returns the closest integer to 'number' that is divisible by 'factor'."""
+    return round(number / factor) * factor
+
+
+def ceil_by_factor(number: int, factor: int) -> int:
+    """Returns the smallest integer greater than or equal to 'number' that is
+    divisible by 'factor'."""
+    return math.ceil(number / factor) * factor
+
+
+def floor_by_factor(number: int, factor: int) -> int:
+    """Returns the largest integer less than or equal to 'number' that is
+    divisible by 'factor'."""
+    return math.floor(number / factor) * factor
+
+
+def smart_resize(
+    height: int,
+    width: int,
+    factor: int = IMAGE_FACTOR,
+    min_pixels: int = MIN_PIXELS,
+    max_pixels: int = MAX_PIXELS,
+) -> tuple[int, int]:
+    """
+    Rescales the image so that the following conditions are met:
+
+    1. Both dimensions (height and width) are divisible by 'factor'.
+
+    2. The total number of pixels is within the range ['min_pixels', 'max_pixels'].
+
+    3. The aspect ratio of the image is maintained as closely as possible.
+    """
+    if max(height, width) / min(height, width) > MAX_RATIO:
+        raise ValueError(
+            f"absolute aspect ratio must be smaller than {MAX_RATIO}, got {max(height, width) / min(height, width)}"
+        )
+    h_bar = max(factor, round_by_factor(height, factor))
+    w_bar = max(factor, round_by_factor(width, factor))
+    if h_bar * w_bar > max_pixels:
+        beta = math.sqrt((height * width) / max_pixels)
+        h_bar = floor_by_factor(height / beta, factor)
+        w_bar = floor_by_factor(width / beta, factor)
+    elif h_bar * w_bar < min_pixels:
+        beta = math.sqrt(min_pixels / (height * width))
+        h_bar = ceil_by_factor(height * beta, factor)
+        w_bar = ceil_by_factor(width * beta, factor)
+    return h_bar, w_bar
+
+
+def fetch_image(ele: dict[str, str | Image.Image], size_factor: int = IMAGE_FACTOR) -> Image.Image:
+    if "image" in ele:
+        image = ele["image"]
+    else:
+        image = ele["image_url"]
+    image_obj = None
+    if isinstance(image, Image.Image):
+        image_obj = image
+    elif image.startswith("http://") or image.startswith("https://"):
+        image_obj = Image.open(requests.get(image, stream=True).raw)
+    elif image.startswith("file://"):
+        image_obj = Image.open(image[7:])
+    elif image.startswith("data:image"):
+        if "base64," in image:
+            _, base64_data = image.split("base64,", 1)
+            data = base64.b64decode(base64_data)
+            image_obj = Image.open(BytesIO(data))
+    else:
+        image_obj = Image.open(image)
+    if image_obj is None:
+        raise ValueError(f"Unrecognized image input, support local path, http url, base64 and PIL.Image, got {image}")
+    image = image_obj.convert("RGB")
+    # resize
+    if "resized_height" in ele and "resized_width" in ele:
+        resized_height, resized_width = smart_resize(
+            ele["resized_height"],
+            ele["resized_width"],
+            factor=size_factor,
+        )
+    else:
+        width, height = image.size
+        min_pixels = ele.get("min_pixels", MIN_PIXELS)
+        max_pixels = ele.get("max_pixels", MAX_PIXELS)
+        resized_height, resized_width = smart_resize(
+            height,
+            width,
+            factor=size_factor,
+            min_pixels=min_pixels,
+            max_pixels=max_pixels,
+        )
+    image = image.resize((resized_width, resized_height))
+
+    return image
+
+
+def smart_nframes(
+    ele: dict,
+    total_frames: int,
+    video_fps: int | float,
+) -> int:
+    """calculate the number of frames for video used for model inputs.
+
+    Args:
+        ele (dict): a dict contains the configuration of video.
+            support either `fps` or `nframes`:
+                - nframes: the number of frames to extract for model inputs.
+                - fps: the fps to extract frames for model inputs.
+                    - min_frames: the minimum number of frames of the video,
+                      only used when fps is provided.
+                    - max_frames: the maximum number of frames of the video,
+                      only used when fps is provided.
+        total_frames (int): the original total number of frames of the video.
+        video_fps (int | float): the original fps of the video.
+
+    Raises:
+        ValueError: nframes should in interval [FRAME_FACTOR, total_frames].
+
+    Returns:
+        int: the number of frames for video used for model inputs.
+    """
+    assert not ("fps" in ele and "nframes" in ele), "Only accept either `fps` or `nframes`"
+    if "nframes" in ele:
+        nframes = round_by_factor(ele["nframes"], FRAME_FACTOR)
+    else:
+        fps = ele.get("fps", FPS)
+        min_frames = ceil_by_factor(ele.get("min_frames", FPS_MIN_FRAMES), FRAME_FACTOR)
+        max_frames = floor_by_factor(
+            ele.get("max_frames", min(FPS_MAX_FRAMES, total_frames)),
+            FRAME_FACTOR,
+        )
+        nframes = total_frames / video_fps * fps
+        nframes = min(max(nframes, min_frames), max_frames)
+        nframes = round_by_factor(nframes, FRAME_FACTOR)
+    if not (FRAME_FACTOR <= nframes and nframes <= total_frames):
+        raise ValueError(f"nframes should in interval [{FRAME_FACTOR}, {total_frames}], but got {nframes}.")
+    return nframes
+
+
+def _read_video_torchvision(
+    ele: dict,
+) -> torch.Tensor:
+    """read video using torchvision.io.read_video
+
+    Args:
+        ele (dict): a dict contains the configuration of video.
+        support keys:
+            - video: the path of video. support "file://", "http://",
+              "https://" and local path.
+            - video_start: the start time of video.
+            - video_end: the end time of video.
+    Returns:
+        torch.Tensor: the video tensor with shape (T, C, H, W).
+    """
+    video_path = ele["video"]
+    if version.parse(torchvision.__version__) < version.parse("0.19.0"):
+        if "http://" in video_path or "https://" in video_path:
+            warnings.warn("torchvision < 0.19.0 does not support http/https video path, please upgrade to 0.19.0.")
+        if "file://" in video_path:
+            video_path = video_path[7:]
+    st = time.time()
+    video, audio, info = io.read_video(
+        video_path,
+        start_pts=ele.get("video_start", 0.0),
+        end_pts=ele.get("video_end", None),
+        pts_unit="sec",
+        output_format="TCHW",
+    )
+    total_frames, video_fps = video.size(0), info["video_fps"]
+    total_duration = round(total_frames / video_fps, 3)
+    logger.info(
+        f"torchvision:  {video_path=}, {total_frames=}, {video_fps=}, duration={total_duration}s, time={time.time() - st:.3f}s"
+    )
+    nframes = smart_nframes(ele, total_frames=total_frames, video_fps=video_fps)
+    idx = torch.linspace(0, total_frames - 1, nframes).round().long()
+    video = video[idx]
+    return video, total_duration, nframes
+
+
+def is_decord_available() -> bool:
+    import importlib.util
+
+    return importlib.util.find_spec("decord") is not None
+
+
+def _read_video_decord(
+    ele: dict,
+) -> torch.Tensor:
+    """read video using decord.VideoReader
+
+    Args:
+        ele (dict): a dict contains the configuration of video.
+        support keys:
+            - video: the path of video. support "file://", "http://",
+              "https://" and local path.
+            - video_start: the start time of video.
+            - video_end: the end time of video.
+    Returns:
+        torch.Tensor: the video tensor with shape (T, C, H, W).
+    """
+    import decord
+
+    video_path = ele["video"]
+    st = time.time()
+    vr = decord.VideoReader(video_path)
+    # TODO: support start_pts and end_pts
+    if "video_start" in ele or "video_end" in ele:
+        raise NotImplementedError("not support start_pts and end_pts in decord for now.")
+    total_frames, video_fps = len(vr), vr.get_avg_fps()
+    total_duration = round(total_frames / video_fps, 3)
+    logger.info(f"decord:  {video_path=}, {total_frames=}, {video_fps=}, time={time.time() - st:.3f}s")
+    nframes = smart_nframes(ele, total_frames=total_frames, video_fps=video_fps)
+    idx = torch.linspace(0, total_frames - 1, nframes).round().long().tolist()
+    video = vr.get_batch(idx).asnumpy()
+    video = torch.tensor(video).permute(0, 3, 1, 2)  # Convert to TCHW format
+    return video, total_duration, nframes
+
+
+VIDEO_READER_BACKENDS = {
+    "decord": _read_video_decord,
+    "torchvision": _read_video_torchvision,
+}
+
+FORCE_QWENVL_VIDEO_READER = os.getenv("FORCE_QWENVL_VIDEO_READER", None)
+
+
+@lru_cache(maxsize=1)
+def get_video_reader_backend() -> str:
+    if FORCE_QWENVL_VIDEO_READER is not None:
+        video_reader_backend = FORCE_QWENVL_VIDEO_READER
+    elif is_decord_available():
+        video_reader_backend = "decord"
+    else:
+        video_reader_backend = "torchvision"
+    # print(f"qwen-vl-utils using {video_reader_backend} to read video.",
+    #       file=sys.stderr)
+    return video_reader_backend
+
+
+def fetch_video(ele: dict, image_factor: int = IMAGE_FACTOR) -> torch.Tensor | list[Image.Image]:
+    if isinstance(ele["video"], str):
+        video_reader_backend = get_video_reader_backend()
+        video, total_dur, nframes = VIDEO_READER_BACKENDS[video_reader_backend](ele)
+        frame_timestamps = total_dur * torch.arange(1, nframes + 1) / nframes
+        grid_timestamps = frame_timestamps[::FRAME_FACTOR]
+        second_per_grid = grid_timestamps[1] - grid_timestamps[0]
+        nframes, _, height, width = video.shape
+        min_pixels = ele.get("min_pixels", VIDEO_MIN_PIXELS)
+        total_pixels = ele.get("total_pixels", VIDEO_TOTAL_PIXELS)
+        max_pixels = max(
+            min(VIDEO_MAX_PIXELS, total_pixels / nframes * FRAME_FACTOR),
+            int(min_pixels * 1.05),
+        )
+        max_pixels = ele.get("max_pixels", max_pixels)
+        # min_pixels = (factor ** 2) * 52
+        # max_pixels = (factor ** 2) * min(768, (16384 / nframes * temporal_patch_size))
+        if "resized_height" in ele and "resized_width" in ele:
+            resized_height, resized_width = smart_resize(
+                ele["resized_height"],
+                ele["resized_width"],
+                factor=image_factor,
+            )
+        else:
+            resized_height, resized_width = smart_resize(
+                height,
+                width,
+                factor=image_factor,
+                min_pixels=min_pixels,
+                max_pixels=max_pixels,
+            )
+        video = transforms.functional.resize(
+            video,
+            [resized_height, resized_width],
+            interpolation=InterpolationMode.BICUBIC,
+            antialias=True,
+        ).float()
+        return video, total_dur, nframes, second_per_grid
+    else:
+        assert isinstance(ele["video"], (list, tuple))
+        process_info = ele.copy()
+        process_info.pop("type", None)
+        process_info.pop("video", None)
+        images = [
+            fetch_image({"image": video_element, **process_info}, size_factor=image_factor)
+            for video_element in ele["video"]
+        ]
+        nframes = ceil_by_factor(len(images), FRAME_FACTOR)
+        if len(images) < nframes:
+            images.extend([images[-1]] * (nframes - len(images)))
+        return images, None, None, None
+
+
+def extract_vision_info(conversations: list[dict] | list[list[dict]]) -> list[dict]:
+    vision_infos = []
+    if isinstance(conversations[0], dict):
+        conversations = [conversations]
+    for conversation in conversations:
+        for message in conversation:
+            if isinstance(message["content"], list):
+                for ele in message["content"]:
+                    if (
+                        "image" in ele
+                        or "image_url" in ele
+                        or "video" in ele
+                        or ele["type"] in ("image", "image_url", "video")
+                    ):
+                        vision_infos.append(ele)
+    return vision_infos
+
+
+def process_vision_info(
+    conversations: list[dict] | list[list[dict]],
+) -> tuple[list[Image.Image] | None, list[torch.Tensor | list[Image.Image]] | None]:
+    vision_infos = extract_vision_info(conversations)
+    # Read images or videos
+    image_inputs = []
+    video_inputs = []
+    for vision_info in vision_infos:
+        if "image" in vision_info or "image_url" in vision_info:
+            image_inputs.append(fetch_image(vision_info))
+        elif "video" in vision_info:
+            video_inputs.append(fetch_video(vision_info))
+        else:
+            raise ValueError("image, image_url or video should in content.")
+    if len(image_inputs) == 0:
+        image_inputs = None
+    if len(video_inputs) == 0:
+        video_inputs = None
+    return image_inputs, video_inputs

examples/offline_inference/qwen2_5_omni/run_multiple_prompts.sh

@@ -1,3 +1,8 @@
-python end2end.py --output-wav output_audio \
-                  --query-type text \
-                  --txt-prompts top10.txt
+python end2end.py --model Qwen/Qwen2.5-Omni-7B \
+                                 --voice-type "m02" \
+                                 --dit-ckpt none \
+                                 --bigvgan-ckpt none \
+                                 --output-wav output_audio \
+                                 --prompt_type text \
+                                 --init-sleep-seconds 0 \
+                                 --txt-prompts top100.txt

examples/offline_inference/qwen2_5_omni/run_single_prompt.sh

@@ -1,2 +1,8 @@
-python end2end.py --output-wav output_audio \
-                  --query-type mixed_modalities
+python end2end.py --model Qwen/Qwen2.5-Omni-7B \
+                                 --voice-type "m02" \
+                                 --dit-ckpt none \
+                                 --bigvgan-ckpt none \
+                                 --output-wav output_audio \
+                                 --prompt_type text \
+                                 --init-sleep-seconds 0 \
+                                 --prompts "Explain the system architecture for a scalable audio generation pipeline. Answer in 15 words."

examples/offline_inference/qwen2_5_omni/utils.py

@@ -0,0 +1,312 @@
+import tempfile
+from typing import Optional, Union
+from urllib.request import urlopen
+
+import librosa
+import requests
+import resampy
+import soundfile as sf
+import torch
+import torchvision.io
+from processing_omni import fetch_image, fetch_video
+from transformers import AutoConfig, AutoProcessor
+from vllm.inputs import TextPrompt
+
+from vllm_omni.inputs.data import OmniTokensPrompt
+
+# Simple caches to avoid repeated heavy HF loads per prompt
+_PROCESSOR_CACHE: dict[str, "AutoProcessor"] = {}
+_CONFIG_CACHE: dict[str, "AutoConfig"] = {}
+
+
+def get_system_prompt():
+    return {
+        "role": "system",
+        "content": [
+            {
+                "type": "text",
+                "text": (
+                    "You are Qwen, a virtual human developed by the Qwen Team, "
+                    "Alibaba Group, capable of perceiving auditory and visual inputs, "
+                    "as well as generating text and speech."
+                ),
+            }
+        ],
+    }
+
+
+def resample_wav_to_16khz(input_filepath):
+    data, original_sample_rate = sf.read(input_filepath)
+    # Only use the first channel
+    if len(data.shape) > 1:
+        data = data[:, 0]
+    # resample to 16kHz
+    data_resampled = resampy.resample(data, sr_orig=original_sample_rate, sr_new=16000)
+    return data_resampled
+
+
+def fetch_and_read_video(args, video_url: str, fps=2):
+    def read_video_with_torchvision(video_file_name: str):
+        video, audio, info = torchvision.io.read_video(
+            video_file_name,
+            start_pts=0.0,
+            end_pts=None,
+            pts_unit="sec",
+            output_format="TCHW",
+        )
+
+        total_frames, video_fps = video.size(0), info["video_fps"]
+        total_duration = round(total_frames / video_fps, 3)
+        nframes = int(total_frames / video_fps * fps)
+
+        frame_timestamps = total_duration * torch.arange(1, nframes + 1) / nframes
+        grid_timestamps = frame_timestamps[::2]
+        second_per_grid = grid_timestamps[1] - grid_timestamps[0]
+
+        idx = torch.linspace(0, video.size(0) - 1, nframes).round().long()
+        video = video[idx]
+
+        if args.legacy_omni_video:
+            return [video, total_duration, nframes, second_per_grid.item()]
+        else:
+            return video
+
+    def read_video_with_transformers(video_file_name: Union[str, list[str]]):
+        video, total_duration, nframes, second_per_grid = fetch_video({"video": video_file_name})
+        if total_duration is None and nframes is None:
+            nframes = len(video)
+            total_duration = 0.5 * nframes
+            second_per_grid = 1.0
+        if args.legacy_omni_video:
+            return [video, total_duration, nframes, second_per_grid]
+        else:
+            return video
+
+    def read_video(video_file_name: str):
+        if args.use_torchvision:
+            return read_video_with_torchvision(video_file_name)
+        else:
+            return read_video_with_transformers(video_file_name)
+
+    if isinstance(video_url, str) and video_url.startswith("http"):
+        with tempfile.NamedTemporaryFile(delete=True) as temp_video_file:
+            resp = requests.get(video_url)
+            assert resp.status_code == requests.codes.ok, (
+                f"Failed to fetch video from {video_url}, status_code:{resp.status_code}, resp:{resp}"
+            )
+
+            temp_video_file.write(urlopen(video_url).read())
+            temp_video_file_path = temp_video_file.name
+            video_file_name = temp_video_file_path
+            return read_video(video_file_name)
+    else:
+        video_file_name = video_url
+        return read_video(video_file_name)
+
+
+def make_inputs_qwen2_omni(
+    args,
+    messages: list[dict[str, Union[str, list[dict[str, str]]]]],
+    use_audio_in_video: Optional[bool] = False,
+    tokenize: bool = False,
+) -> Union[OmniTokensPrompt, TextPrompt]:
+    from transformers import AutoConfig, AutoProcessor
+
+    # Cached processor/config to prevent per-prompt reloading and repeated warnings
+    if args.model not in _PROCESSOR_CACHE:
+        _PROCESSOR_CACHE[args.model] = AutoProcessor.from_pretrained(args.model)
+    processor = _PROCESSOR_CACHE[args.model]
+
+    config = _CONFIG_CACHE.get(args.model)
+    if config is None:
+        try:
+            config = AutoConfig.from_pretrained(args.model)
+        except Exception:
+            config = None
+        _CONFIG_CACHE[args.model] = config  # cache even if None to avoid retry storms
+
+    # Decide legacy flag only once based on config (default True if unknown)
+    if getattr(args, "legacy_omni_video", None) is None:
+        if config is not None and hasattr(config, "architectures"):
+            args.legacy_omni_video = "Qwen2_5OmniModel" not in config.architectures
+        else:
+            args.legacy_omni_video = True
+
+    audios, images, videos = [], [], []
+    for message in messages:
+        if not isinstance(message["content"], list):
+            message["content"] = [
+                {
+                    "type": "text",
+                    "text": message["content"],
+                }
+            ]
+        index, num_contents = 0, len(message["content"])
+        while index < num_contents:
+            ele = message["content"][index]
+            if "type" not in ele:
+                if "text" in ele:
+                    ele["type"] = "text"
+                elif "audio" in ele:
+                    ele["type"] = "audio"
+                elif "audio_url" in ele:
+                    ele["type"] = "audio_url"
+                elif "image" in ele:
+                    ele["type"] = "image"
+                elif "image_url" in ele:
+                    ele["type"] = "image_url"
+                elif "video" in ele:
+                    ele["type"] = "video"
+                elif "video_url" in ele:
+                    ele["type"] = "video_url"
+                else:
+                    raise ValueError(f"Unknown ele: {ele}")
+
+            if ele["type"] == "audio" or ele["type"] == "audio_url":
+                if "audio_url" in ele:
+                    audio_key = "audio_url"
+                    with tempfile.NamedTemporaryFile(delete=True) as temp_audio_file:
+                        temp_audio_file.write(urlopen(ele[audio_key]).read())
+                        temp_audio_file_path = temp_audio_file.name
+                        audios.append(resample_wav_to_16khz(temp_audio_file_path))
+                        ele["audio"] = temp_audio_file_path
+                elif "audio" in ele:
+                    audio_key = "audio"
+                    audios.append(resample_wav_to_16khz(ele[audio_key]))
+                else:
+                    raise ValueError(f"Unknown ele {ele}")
+            elif use_audio_in_video and (ele["type"] == "video" or ele["type"] == "video_url"):
+                # use video as audio as well
+                if "video_url" in ele:
+                    audio_key = "video_url"
+                    with tempfile.NamedTemporaryFile(delete=True) as temp_video_file:
+                        temp_video_file.write(urlopen(ele[audio_key]).read())
+                        temp_video_file_path = temp_video_file.name
+                        ele[audio_key] = temp_video_file_path
+                        audios.append(librosa.load(temp_video_file_path, sr=16000)[0])
+                        videos.append(fetch_and_read_video(args, temp_video_file_path))
+                        ele["video"] = temp_video_file_path
+                elif "video" in ele:
+                    audio_key = "video"
+                    audios.append(librosa.load(ele[audio_key], sr=16000)[0])
+                    videos.append(fetch_and_read_video(args, audio_key))
+                else:
+                    raise ValueError(f"Unknown ele {ele}")
+                # insert a audio after the video
+                message["content"].insert(
+                    index + 1,
+                    {
+                        "type": "audio",
+                        "audio": ele[audio_key],
+                    },
+                )
+                # no need to load the added audio again
+                index += 1
+            elif ele["type"] == "video" or ele["type"] == "video_url":
+                if "video_url" in ele:
+                    video_key = "video_url"
+                    with tempfile.NamedTemporaryFile(delete=True) as temp_video_file:
+                        temp_video_file.write(urlopen(ele["video_url"]).read())
+                        temp_video_file_path = temp_video_file.name
+                        videos.append(fetch_and_read_video(args, temp_video_file))
+                        ele["video"] = temp_video_file_path
+                else:
+                    video_key = "video"
+                    videos.append(fetch_and_read_video(args, ele[video_key]))
+            elif ele["type"] == "image" or ele["type"] == "image_url":
+                images.append(fetch_image(ele))
+
+            # move to the next content
+            index += 1
+
+    prompt = processor.apply_chat_template(
+        messages,
+        tokenize=tokenize,
+        add_generation_prompt=True,
+        add_vision_id=True,
+    )
+
+    audios = audios if len(audios) > 0 else None
+    images = images if len(images) > 0 else None
+    videos = videos if len(videos) > 0 else None
+
+    multi_modal_data = {}
+    if audios:
+        multi_modal_data["audio"] = audios
+    if images:
+        multi_modal_data["image"] = images
+    if videos:
+        multi_modal_data["video"] = videos
+
+    if isinstance(prompt, list) and isinstance(prompt[0], (list, str)):
+        prompt = prompt[0]
+
+    if tokenize:
+        return OmniTokensPrompt(
+            prompt_token_ids=prompt,
+            multi_modal_data=multi_modal_data,
+        )
+    else:
+        return TextPrompt(
+            prompt=prompt,
+            multi_modal_data=multi_modal_data,
+        )
+
+
+def make_text_prompt(args, prompt):
+    messages = [
+        get_system_prompt(),
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": prompt},
+            ],
+        },
+    ]
+
+    prompt = make_inputs_qwen2_omni(args, messages, tokenize=args.tokenize)
+    return prompt
+
+
+def make_audio_in_video_v2_prompt(args):
+    messages = [
+        {
+            "role": "system",
+            "content": [
+                {
+                    "type": "text",
+                    "text": (
+                        "You are Qwen, a virtual human developed by the Qwen Team, "
+                        "Alibaba Group, capable of perceiving auditory and visual "
+                        "inputs, as well as generating text and speech."
+                    ),
+                }
+            ],
+        },
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "video_url",
+                    "video_url": ("https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2.5-Omni/draw_small.mp4"),
+                },
+            ],
+        },
+    ]
+    prompt = make_inputs_qwen2_omni(
+        args,
+        messages,
+        use_audio_in_video=True,
+        tokenize=args.tokenize,
+    )
+    return prompt
+
+
+def make_omni_prompt(args, prompt=None) -> Union[OmniTokensPrompt, list[OmniTokensPrompt]]:
+    if args.prompt_type == "text":
+        prompt = make_text_prompt(args, prompt)
+    elif args.prompt_type == "audio-in-video-v2":
+        prompt = make_audio_in_video_v2_prompt(args)
+    else:
+        raise ValueError(f"Unsupported prompt type: {args.prompt_type}")
+    return prompt

examples/offline_inference/qwen3_omni/README.md

@@ -1,59 +0,0 @@
-# Offline Example of vLLM-Omni for Qwen3-Omni
-
-## 🛠️ Installation
-
-Please refer to [README.md](../../../README.md)
-
-## Run examples (Qwen3-Omni)
-
-### Multiple Prompts
-Download dataset from [seed_tts](https://drive.google.com/file/d/1GlSjVfSHkW3-leKKBlfrjuuTGqQ_xaLP/edit). For processing dataset please refer to [Qwen2.5-Omni README.md](../qwen2_5_omni/README.md)
-Get into the example folder
-```bash
-cd examples/offline_inference/qwen3_omni
-```
-Then run the command below.
-```bash
-bash run_multiple_prompts.sh
-```
-### Single Prompt
-Get into the example folder
-```bash
-cd examples/offline_inference/qwen3_omni
-```
-Then run the command below.
-```bash
-bash run_single_prompt.sh
-```
-If you have not enough memory, you can set thinker with tensor parallel. Just run the command below.
-```bash
-bash run_single_prompt_tp.sh
-```
-
-#### Using Local Media Files
-The `end2end.py` script supports local media files (audio, video, image) via command-line arguments:
-
-```bash
-# Use local video file
-python end2end.py --query-type use_video --video-path /path/to/video.mp4
-
-# Use local image file
-python end2end.py --query-type use_image --image-path /path/to/image.jpg
-
-# Use local audio file
-python end2end.py --query-type use_audio --audio-path /path/to/audio.wav
-```
-
-If media file paths are not provided, the script will use default assets. Supported query types:
-- `use_video`: Video input
-- `use_image`: Image input
-- `use_audio`: Audio input
-- `text`: Text-only query
-
-### FAQ
-
-If you encounter error about backend of librosa, try to install ffmpeg with command below.
-```
-sudo apt update
-sudo apt install ffmpeg
-```

examples/offline_inference/qwen3_omni/end2end.py

@@ -1,368 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""
-This example shows how to use vLLM for running offline inference
-with the correct prompt format on Qwen3-Omni (thinker only).
-"""
-
-import os
-from typing import NamedTuple, Optional
-
-import librosa
-import numpy as np
-import soundfile as sf
-from PIL import Image
-from vllm import SamplingParams
-from vllm.assets.audio import AudioAsset
-from vllm.assets.image import ImageAsset
-from vllm.assets.video import VideoAsset, video_to_ndarrays
-from vllm.multimodal.image import convert_image_mode
-from vllm.utils import FlexibleArgumentParser
-
-from vllm_omni.entrypoints.omni import Omni
-
-SEED = 42
-
-
-class QueryResult(NamedTuple):
-    inputs: dict
-    limit_mm_per_prompt: dict[str, int]
-
-
-# NOTE: The default `max_num_seqs` and `max_model_len` may result in OOM on
-# lower-end GPUs.
-# Unless specified, these settings have been tested to work on a single L4.
-
-default_system = (
-    "You are Qwen, a virtual human developed by the Qwen Team, Alibaba "
-    "Group, capable of perceiving auditory and visual inputs, as well as "
-    "generating text and speech."
-)
-
-
-def get_text_query(question: str = None) -> QueryResult:
-    if question is None:
-        question = "Explain the system architecture for a scalable audio generation pipeline. Answer in 15 words."
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
-    )
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-        },
-        limit_mm_per_prompt={},
-    )
-
-
-def get_video_query(question: str = None, video_path: Optional[str] = None, num_frames: int = 16) -> QueryResult:
-    if question is None:
-        question = "Why is this video funny?"
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
-    )
-
-    if video_path:
-        if not os.path.exists(video_path):
-            raise FileNotFoundError(f"Video file not found: {video_path}")
-        video_frames = video_to_ndarrays(video_path, num_frames=num_frames)
-    else:
-        video_frames = VideoAsset(name="baby_reading", num_frames=num_frames).np_ndarrays
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "video": video_frames,
-            },
-        },
-        limit_mm_per_prompt={"video": 1},
-    )
-
-
-def get_image_query(question: str = None, image_path: Optional[str] = None) -> QueryResult:
-    if question is None:
-        question = "What is the content of this image?"
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
-    )
-
-    if image_path:
-        if not os.path.exists(image_path):
-            raise FileNotFoundError(f"Image file not found: {image_path}")
-        pil_image = Image.open(image_path)
-        image_data = convert_image_mode(pil_image, "RGB")
-    else:
-        image_data = convert_image_mode(ImageAsset("cherry_blossom").pil_image, "RGB")
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "image": image_data,
-            },
-        },
-        limit_mm_per_prompt={"image": 1},
-    )
-
-
-def get_audio_query(question: str = None, audio_path: Optional[str] = None, sampling_rate: int = 16000) -> QueryResult:
-    if question is None:
-        question = "What is the content of this audio?"
-    prompt = (
-        f"<|im_start|>system\n{default_system}<|im_end|>\n"
-        "<|im_start|>user\n<|audio_start|><|audio_pad|><|audio_end|>"
-        f"{question}<|im_end|>\n"
-        f"<|im_start|>assistant\n"
-    )
-
-    if audio_path:
-        if not os.path.exists(audio_path):
-            raise FileNotFoundError(f"Audio file not found: {audio_path}")
-        audio_signal, sr = librosa.load(audio_path, sr=sampling_rate)
-        audio_data = (audio_signal.astype(np.float32), sr)
-    else:
-        audio_data = AudioAsset("mary_had_lamb").audio_and_sample_rate
-
-    return QueryResult(
-        inputs={
-            "prompt": prompt,
-            "multi_modal_data": {
-                "audio": audio_data,
-            },
-        },
-        limit_mm_per_prompt={"audio": 1},
-    )
-
-
-query_map = {
-    "text": get_text_query,
-    "use_audio": get_audio_query,
-    "use_image": get_image_query,
-    "use_video": get_video_query,
-}
-
-
-def main(args):
-    model_name = "Qwen/Qwen3-Omni-30B-A3B-Instruct"
-
-    # Get paths from args
-    video_path = getattr(args, "video_path", None)
-    image_path = getattr(args, "image_path", None)
-    audio_path = getattr(args, "audio_path", None)
-
-    # Get the query function and call it with appropriate parameters
-    query_func = query_map[args.query_type]
-    if args.query_type == "use_video":
-        query_result = query_func(video_path=video_path, num_frames=getattr(args, "num_frames", 16))
-    elif args.query_type == "use_image":
-        query_result = query_func(image_path=image_path)
-    elif args.query_type == "use_audio":
-        query_result = query_func(audio_path=audio_path, sampling_rate=getattr(args, "sampling_rate", 16000))
-    else:
-        query_result = query_func()
-
-    omni_llm = Omni(
-        model=model_name,
-        stage_configs_path=args.stage_configs_path,
-    )
-
-    thinker_sampling_params = SamplingParams(
-        temperature=0.4,
-        top_p=0.9,
-        top_k=-1,
-        max_tokens=1200,
-        repetition_penalty=1.05,
-        logit_bias={},
-        seed=SEED,
-    )
-
-    talker_sampling_params = SamplingParams(
-        temperature=0.9,
-        top_k=50,
-        max_tokens=4096,
-        seed=SEED,
-        detokenize=False,
-        repetition_penalty=1.05,
-        stop_token_ids=[2150],  # TALKER_CODEC_EOS_TOKEN_ID
-    )
-
-    # Sampling parameters for Code2Wav stage (audio generation)
-    code2wav_sampling_params = SamplingParams(
-        temperature=0.0,
-        top_p=1.0,
-        top_k=-1,
-        max_tokens=4096 * 16,
-        seed=SEED,
-        detokenize=True,
-        repetition_penalty=1.1,
-    )
-
-    sampling_params_list = [
-        thinker_sampling_params,
-        talker_sampling_params,  # code predictor is integrated into talker for Qwen3 Omni
-        code2wav_sampling_params,
-    ]
-
-    if args.txt_prompts is None:
-        prompts = [query_result.inputs for _ in range(args.num_prompts)]
-    else:
-        assert args.query_type == "text", "txt-prompts is only supported for text query type"
-        with open(args.txt_prompts, encoding="utf-8") as f:
-            lines = [ln.strip() for ln in f.readlines()]
-            prompts = [get_text_query(ln).inputs for ln in lines if ln != ""]
-            print(f"[Info] Loaded {len(prompts)} prompts from {args.txt_prompts}")
-
-    omni_outputs = omni_llm.generate(prompts, sampling_params_list)
-    # Determine output directory: prefer --output-dir; fallback to --output-wav
-    output_dir = args.output_dir if getattr(args, "output_dir", None) else args.output_wav
-    os.makedirs(output_dir, exist_ok=True)
-
-    for stage_outputs in omni_outputs:
-        if stage_outputs.final_output_type == "text":
-            for output in stage_outputs.request_output:
-                request_id = int(output.request_id)
-                text_output = output.outputs[0].text
-                # Save aligned text file per request
-                prompt_text = prompts[request_id]["prompt"]
-                out_txt = os.path.join(output_dir, f"{request_id:05d}.txt")
-                lines = []
-                lines.append("Prompt:\n")
-                lines.append(str(prompt_text) + "\n")
-                lines.append("vllm_text_output:\n")
-                lines.append(str(text_output).strip() + "\n")
-                try:
-                    with open(out_txt, "w", encoding="utf-8") as f:
-                        f.writelines(lines)
-                except Exception as e:
-                    print(f"[Warn] Failed writing text file {out_txt}: {e}")
-                print(f"Request ID: {request_id}, Text saved to {out_txt}")
-        elif stage_outputs.final_output_type == "audio":
-            for output in stage_outputs.request_output:
-                request_id = int(output.request_id)
-                audio_tensor = output.multimodal_output["audio"]
-                output_wav = os.path.join(output_dir, f"output_{output.request_id}.wav")
-
-                # Convert to numpy array and ensure correct format
-                audio_numpy = audio_tensor.float().detach().cpu().numpy()
-
-                # Ensure audio is 1D (flatten if needed)
-                if audio_numpy.ndim > 1:
-                    audio_numpy = audio_numpy.flatten()
-
-                # Save audio file with explicit WAV format
-                sf.write(output_wav, audio_numpy, samplerate=24000, format="WAV")
-                print(f"Request ID: {request_id}, Saved audio to {output_wav}")
-
-
-def parse_args():
-    parser = FlexibleArgumentParser(description="Demo on using vLLM for offline inference with audio language models")
-    parser.add_argument(
-        "--query-type",
-        "-q",
-        type=str,
-        default="mixed_modalities",
-        choices=query_map.keys(),
-        help="Query type.",
-    )
-    parser.add_argument(
-        "--enable-stats",
-        action="store_true",
-        default=False,
-        help="Enable writing detailed statistics (default: disabled)",
-    )
-    parser.add_argument(
-        "--init-sleep-seconds",
-        type=int,
-        default=20,
-        help="Sleep seconds after starting each stage process to allow initialization (default: 20)",
-    )
-    parser.add_argument(
-        "--batch-timeout",
-        type=int,
-        default=5,
-        help="Timeout for batching in seconds (default: 5)",
-    )
-    parser.add_argument(
-        "--init-timeout",
-        type=int,
-        default=300,
-        help="Timeout for initializing stages in seconds (default: 300)",
-    )
-    parser.add_argument(
-        "--shm-threshold-bytes",
-        type=int,
-        default=65536,
-        help="Threshold for using shared memory in bytes (default: 65536)",
-    )
-    parser.add_argument(
-        "--output-wav",
-        default="output_audio",
-        help="[Deprecated] Output wav directory (use --output-dir).",
-    )
-    parser.add_argument(
-        "--num-prompts",
-        type=int,
-        default=1,
-        help="Number of prompts to generate.",
-    )
-    parser.add_argument(
-        "--txt-prompts",
-        type=str,
-        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.",
-    )
-    parser.add_argument(
-        "--video-path",
-        "-v",
-        type=str,
-        default=None,
-        help="Path to local video file. If not provided, uses default video asset.",
-    )
-    parser.add_argument(
-        "--image-path",
-        "-i",
-        type=str,
-        default=None,
-        help="Path to local image file. If not provided, uses default image asset.",
-    )
-    parser.add_argument(
-        "--audio-path",
-        "-a",
-        type=str,
-        default=None,
-        help="Path to local audio file. If not provided, uses default audio asset.",
-    )
-    parser.add_argument(
-        "--num-frames",
-        type=int,
-        default=16,
-        help="Number of frames to extract from video (default: 16).",
-    )
-    parser.add_argument(
-        "--sampling-rate",
-        type=int,
-        default=16000,
-        help="Sampling rate for audio loading (default: 16000).",
-    )
-
-    return parser.parse_args()
-
-
-if __name__ == "__main__":
-    args = parse_args()
-    main(args)

examples/offline_inference/qwen3_omni/run_multiple_prompts.sh

@@ -1,3 +0,0 @@
-python end2end.py --output-wav output_audio \
-                  --query-type text \
-                  --txt-prompts top10.txt

examples/offline_inference/qwen3_omni/run_single_prompt.sh

@@ -1,2 +0,0 @@
-python end2end.py --output-wav output_audio \
-                  --query-type use_audio

examples/offline_inference/qwen3_omni/run_single_prompt_tp.sh

@@ -1,5 +0,0 @@
-python end2end.py --output-wav output_audio \
-                  --query-type use_audio \
-                  --init-sleep-seconds 90
-
-# init-sleep-seconds works to avoid two vLLM stages initialized at the same time within a card.

examples/offline_inference/text_to_image/README.md

@@ -1,54 +0,0 @@
-# Qwen-Image Offline Inference
-
-This folder provides several entrypoints for experimenting with `Qwen/Qwen-Image` using vLLM-Omni:
-
-- `text_to_image.py`: command-line script for single image generation with advanced options.
-- `web_demo.py`: lightweight Gradio UI for interactive prompt/seed/CFG exploration.
-
-## Basic Usage
-
-```python
-from vllm_omni.entrypoints.omni import Omni
-
-if __name__ == "__main__":
-    omni = Omni(model="Qwen/Qwen-Image")
-    prompt = "a cup of coffee on the table"
-    images = omni.generate(prompt)
-    images[0].save("coffee.png")
-```
-
-## Local CLI Usage
-
-```bash
-python text_to_image.py \
-  --prompt "a cup of coffee on the table" \
-  --seed 42 \
-  --cfg_scale 4.0 \
-  --num_images_per_prompt 1 \
-  --num_inference_steps 50 \
-  --height 1024 \
-  --width 1024 \
-  --output outputs/coffee.png
-```
-
-Key arguments:
-
-- `--prompt`: text description (string).
-- `--seed`: integer seed for deterministic sampling.
-- `--cfg_scale`: true CFG scale (model-specific guidance strength).
-- `--num_images_per_prompt`: number of images to generate per prompt (saves as `output`, `output_1`, ...).
-- `--num_inference_steps`: diffusion sampling steps (more steps = higher quality, slower).
-- `--height/--width`: output resolution (defaults 1024x1024).
-- `--output`: path to save the generated PNG.
-
-> ℹ️ Qwen-Image currently publishes best-effort presets at `1328x1328`, `1664x928`, `928x1664`, `1472x1140`, `1140x1472`, `1584x1056`, and `1056x1584`. Adjust `--height/--width` accordingly for the most reliable outcomes.
-
-## Web UI Demo
-
-Launch the gradio demo:
-
-```bash
-python gradio_demo.py --port 7862
-```
-
-Then open `http://localhost:7862/` on your local browser to interact with the web UI.

examples/offline_inference/text_to_image/gradio_demo.py

@@ -1,221 +0,0 @@
-import argparse
-from functools import lru_cache
-
-import gradio as gr
-import torch
-
-from vllm_omni.entrypoints.omni import Omni
-from vllm_omni.utils.platform_utils import detect_device_type, is_npu
-
-ASPECT_RATIOS: dict[str, tuple[int, int]] = {
-    "1:1": (1328, 1328),
-    "16:9": (1664, 928),
-    "9:16": (928, 1664),
-    "4:3": (1472, 1140),
-    "3:4": (1140, 1472),
-    "3:2": (1584, 1056),
-    "2:3": (1056, 1584),
-}
-ASPECT_RATIO_CHOICES = [f"{ratio} ({w}x{h})" for ratio, (w, h) in ASPECT_RATIOS.items()]
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="Gradio demo for Qwen-Image offline inference.")
-    parser.add_argument("--model", default="Qwen/Qwen-Image", help="Diffusion model name or local path.")
-    parser.add_argument(
-        "--height",
-        type=int,
-        default=1328,
-        help="Default image height (must match one of the supported presets).",
-    )
-    parser.add_argument(
-        "--width",
-        type=int,
-        default=1328,
-        help="Default image width (must match one of the supported presets).",
-    )
-    parser.add_argument("--default-prompt", default="a cup of coffee on the table", help="Initial prompt shown in UI.")
-    parser.add_argument("--default-seed", type=int, default=42, help="Initial seed shown in UI.")
-    parser.add_argument("--default-cfg-scale", type=float, default=4.0, help="Initial CFG scale shown in UI.")
-    parser.add_argument(
-        "--num_inference_steps",
-        type=int,
-        default=50,
-        help="Default number of denoising steps shown in the UI.",
-    )
-    parser.add_argument("--ip", default="127.0.0.1", help="Host/IP for Gradio `launch`.")
-    parser.add_argument("--port", type=int, default=7862, help="Port for Gradio `launch`.")
-    parser.add_argument("--share", action="store_true", help="Share the Gradio demo publicly.")
-    args = parser.parse_args()
-    args.aspect_ratio_label = next(
-        (ratio for ratio, dims in ASPECT_RATIOS.items() if dims == (args.width, args.height)),
-        None,
-    )
-    if args.aspect_ratio_label is None:
-        supported = ", ".join(f"{ratio} ({w}x{h})" for ratio, (w, h) in ASPECT_RATIOS.items())
-        parser.error(f"Unsupported resolution {args.width}x{args.height}. Please pick one of: {supported}.")
-    return args
-
-
-@lru_cache(maxsize=1)
-def get_omni(model_name: str) -> Omni:
-    # Enable VAE memory optimizations on NPU
-    vae_use_slicing = is_npu()
-    vae_use_tiling = is_npu()
-    return Omni(
-        model=model_name,
-        vae_use_slicing=vae_use_slicing,
-        vae_use_tiling=vae_use_tiling,
-    )
-
-
-def build_demo(args: argparse.Namespace) -> gr.Blocks:
-    device = detect_device_type()
-    omni = get_omni(args.model)
-
-    def run_inference(
-        prompt: str,
-        seed_value: float,
-        cfg_scale_value: float,
-        resolution_choice: str,
-        num_steps_value: float,
-        num_images_choice: float,
-    ):
-        if not prompt or not prompt.strip():
-            raise gr.Error("Please enter a non-empty prompt.")
-        ratio_label = resolution_choice.split(" ", 1)[0]
-        if ratio_label not in ASPECT_RATIOS:
-            raise gr.Error(f"Unsupported aspect ratio: {ratio_label}")
-        width, height = ASPECT_RATIOS[ratio_label]
-        try:
-            seed = int(seed_value)
-            num_steps = int(num_steps_value)
-            num_images = int(num_images_choice)
-        except (TypeError, ValueError) as exc:
-            raise gr.Error("Seed, inference steps, and number of images must be valid integers.") from exc
-        if num_steps <= 0:
-            raise gr.Error("Inference steps must be a positive integer.")
-        if num_images not in {1, 2, 3, 4}:
-            raise gr.Error("Number of images must be 1, 2, 3, or 4.")
-        generator = torch.Generator(device=device).manual_seed(seed)
-        images = omni.generate(
-            prompt.strip(),
-            height=height,
-            width=width,
-            generator=generator,
-            true_cfg_scale=float(cfg_scale_value),
-            num_inference_steps=num_steps,
-            num_outputs_per_prompt=num_images,
-        )
-        return [img for img in images[:num_images]]
-
-    with gr.Blocks(
-        title="vLLM-Omni Web Serving Demo",
-        css="""
-        /* Left column button width */
-        .left-column button {
-            width: 100%;
-        }
-        /* Right preview area: fixed height, hide unnecessary buttons */
-        .fixed-image {
-            height: 660px;
-            display: flex;
-            flex-direction: column;
-            justify-content: center;
-            align-items: center;
-        }
-        .fixed-image .duplicate-button,
-        .fixed-image .svelte-drgfj2 {
-            display: none !important;
-        }
-        /* Gallery container: fill available space and center content */
-        #image-gallery {
-            width: 100%;
-            height: 100%;
-            display: flex;
-            align-items: center;
-            justify-content: center;
-        }
-        /* Gallery grid: center horizontally and vertically, set gap */
-        #image-gallery .grid {
-            display: flex;
-            flex-wrap: wrap;
-            justify-content: center;
-            align-items: center;
-            align-content: center;
-            gap: 16px;
-            width: 100%;
-            height: 100%;
-        }
-        /* Gallery grid items: center content */
-        #image-gallery .grid > div {
-            display: flex;
-            align-items: center;
-            justify-content: center;
-        }
-        /* Gallery images: limit max height, maintain aspect ratio */
-        .fixed-image img {
-            max-height: 660px !important;
-            width: auto !important;
-            object-fit: contain;
-        }
-        """,
-    ) as demo:
-        gr.Markdown("# vLLM-Omni Web Serving Demo")
-        gr.Markdown(f"**Model:** {args.model}")
-
-        with gr.Row():
-            with gr.Column(scale=1, elem_classes="left-column"):
-                prompt_input = gr.Textbox(
-                    label="Prompt",
-                    value=args.default_prompt,
-                    placeholder="Describe the image you want to generate...",
-                    lines=5,
-                )
-                seed_input = gr.Number(label="Seed", value=args.default_seed, precision=0)
-                cfg_input = gr.Number(label="CFG Scale", value=args.default_cfg_scale)
-                steps_input = gr.Number(
-                    label="Inference Steps",
-                    value=args.num_inference_steps,
-                    precision=0,
-                    minimum=1,
-                )
-                aspect_dropdown = gr.Dropdown(
-                    label="Aspect Ratio (W:H)",
-                    choices=ASPECT_RATIO_CHOICES,
-                    value=f"{args.aspect_ratio_label} ({ASPECT_RATIOS[args.aspect_ratio_label][0]}x{ASPECT_RATIOS[args.aspect_ratio_label][1]})",
-                )
-                num_images = gr.Dropdown(
-                    label="Number of images",
-                    choices=["1", "2", "3", "4"],
-                    value="1",
-                )
-                generate_btn = gr.Button("Generate", variant="primary")
-            with gr.Column(scale=2, elem_classes="fixed-image"):
-                gallery = gr.Gallery(
-                    label="Preview",
-                    columns=2,
-                    rows=2,
-                    height=660,
-                    allow_preview=True,
-                    show_label=True,
-                    elem_id="image-gallery",
-                )
-
-        generate_btn.click(
-            fn=run_inference,
-            inputs=[prompt_input, seed_input, cfg_input, aspect_dropdown, steps_input, num_images],
-            outputs=gallery,
-        )
-
-    return demo
-
-
-def main():
-    args = parse_args()
-    demo = build_demo(args)
-    demo.launch(server_name=args.ip, server_port=args.port, share=args.share)
-
-
-if __name__ == "__main__":
-    main()