fix tensor type hint

.buildkite/pipeline.yml

@@ -1,196 +0,0 @@
-env:
-  IMAGE_VERSION: "py3.12-latest"
-  BUILDKITE_CLEAN_CHECKOUT: true
-
-steps:
-  - label: "pre-commit"
-    command: ".buildkite/scripts/pre_commit.sh"
-    agents:
-      queue: "default"
-
-  - wait
-
-  - label: "Trigger Tests"
-    plugins:
-      - monorepo-diff#v1.4.0:
-          diff: 'git fetch origin "$BUILDKITE_PULL_REQUEST_BASE_BRANCH" && git diff --name-only origin/"$BUILDKITE_PULL_REQUEST_BASE_BRANCH"...HEAD'
-          watch:
-            - path: 
-                - "fastvideo/models/encoders/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/tests/encoders/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 20m .buildkite/scripts/pr_test.sh"
-                label: "Encoder Tests"
-                env:
-                  - TEST_TYPE=encoder
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/models/vaes/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/tests/vaes/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 20m .buildkite/scripts/pr_test.sh"
-                label: "VAE Tests"
-                env:
-                  - TEST_TYPE=vae
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/models/dits/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/tests/transformers/**"
-                - "fastvideo/layers/**"
-                - "fastvideo/attention/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Transformer Tests"
-                env:
-                  - TEST_TYPE=transformer
-                agents:
-                  queue: "default"
-            - path: 
-                - "fastvideo/**/*.py"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 60m .buildkite/scripts/pr_test.sh"
-                label: "SSIM Tests"
-                env:
-                  - TEST_TYPE=ssim
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/tests/lora/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/tests/transformers/**"
-                - "fastvideo/pipelines/**"
-                - "fastvideo/layers/lora/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 20m .buildkite/scripts/pr_test.sh"
-                label: "LoRA Inference Tests"
-                env:
-                  - TEST_TYPE=inference_lora
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Training Tests"
-                env:
-                  - TEST_TYPE=training
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/training/*distillation_pipeline.py"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Distillation DMDTests"
-                env:
-                  - TEST_TYPE=distillation_dmd
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/training/*self_forcing_distillation_pipeline.py"
-                - "fastvideo/tests/training/self-forcing/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Self-Forcing Tests"
-                env:
-                  - TEST_TYPE=self_forcing
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "LoRA Training Tests"
-                env:
-                  - TEST_TYPE=training_lora
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - "fastvideo-kernel/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Training Tests VSA"
-                env:
-                  - TEST_TYPE=training_vsa
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - "fastvideo-kernel/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Inference Tests STA"
-                env:
-                  - TEST_TYPE=inference_sta
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo-kernel/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Kernel Tests"
-                env:
-                  - TEST_TYPE=kernel_tests
-            - path:
-                - "fastvideo-kernel/**"
-                - "fastvideo/attention/backends/vmoba.py"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Inference Tests VMoBA"
-                env:  
-                  - TEST_TYPE=inference_vmoba
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile.python3.12"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: "Unit Tests"
-                env:
-                  - TEST_TYPE=unit_test
-                agents:
-                  queue: "default"
-            # - path:
-            #     - "scripts/lora_extraction/**"
-            #     - "pyproject.toml"
-            #     - "docker/Dockerfile.python3.12"
-            #   config:
-            #     command: "timeout 90m .buildkite/scripts/pr_test.sh"
-            #     label: "LoRA Extraction Tests"
-            #     env:
-            #       - TEST_TYPE=lora_extraction
-            #     agents:
-            #       queue: "default"

.buildkite/scripts/pr_test.sh

@@ -1,142 +0,0 @@
-#!/bin/bash
-set -uo pipefail
-
-log() {
-  echo "[$(date '+%Y-%m-%d %H:%M:%S')] $1"
-}
-
-log "=== Starting Modal test execution ==="
-
-# Change to the project directory
-cd "$(dirname "$0")/../.."
-PROJECT_ROOT=$(pwd)
-log "Project root: $PROJECT_ROOT"
-
-# Install Modal if not available
-if ! python3 -m modal --version &> /dev/null; then
-    log "Modal not found, installing..."
-    python3 -m pip install modal
-    
-    # Verify installation
-    if ! python3 -m modal --version &> /dev/null; then
-        log "Error: Failed to install modal. Please install it manually."
-        exit 1
-    fi
-fi
-
-log "modal version: $(python3 -m modal --version)"
-
-# Set up Modal authentication using Buildkite secrets
-log "Setting up Modal authentication from Buildkite secrets..."
-MODAL_TOKEN_ID=$(buildkite-agent secret get modal_token_id)
-MODAL_TOKEN_SECRET=$(buildkite-agent secret get modal_token_secret)
-
-# Retrieve other secrets
-WANDB_API_KEY=$(buildkite-agent secret get wandb_api_key)
-HF_API_KEY=$(buildkite-agent secret get hf_api_key)
-
-if [ -n "$MODAL_TOKEN_ID" ] && [ -n "$MODAL_TOKEN_SECRET" ]; then
-    log "Retrieved Modal credentials from Buildkite secrets"
-    python3 -m modal token set --token-id "$MODAL_TOKEN_ID" --token-secret "$MODAL_TOKEN_SECRET" --profile buildkite-ci --activate --verify
-    if [ $? -eq 0 ]; then
-        log "Modal authentication successful"
-    else
-        log "Error: Failed to set Modal credentials"
-        exit 1
-    fi
-else
-    log "Error: Could not retrieve Modal credentials from Buildkite secrets."
-    log "Please ensure 'modal_token_id' and 'modal_token_secret' secrets are set in Buildkite."
-    exit 1
-fi
-
-MODAL_TEST_FILE="fastvideo/tests/modal/pr_test.py"
-
-if [ -z "${TEST_TYPE:-}" ]; then
-    log "Error: TEST_TYPE environment variable is not set"
-    exit 1
-fi
-log "Test type: $TEST_TYPE"
-
-MODAL_ENV="BUILDKITE_REPO=$BUILDKITE_REPO BUILDKITE_COMMIT=$BUILDKITE_COMMIT BUILDKITE_PULL_REQUEST=$BUILDKITE_PULL_REQUEST IMAGE_VERSION=$IMAGE_VERSION"
-
-case "$TEST_TYPE" in
-    "encoder")
-        log "Running encoder tests..."
-        MODAL_COMMAND="$MODAL_ENV HF_API_KEY=$HF_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_encoder_tests"
-        ;;
-    "vae")
-        log "Running VAE tests..."
-        MODAL_COMMAND="$MODAL_ENV HF_API_KEY=$HF_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_vae_tests"
-        ;;
-    "transformer")
-        log "Running transformer tests..."
-        MODAL_COMMAND="$MODAL_ENV HF_API_KEY=$HF_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_transformer_tests"
-        ;;
-    "ssim")
-        log "Running SSIM tests..."
-        MODAL_COMMAND="$MODAL_ENV HF_API_KEY=$HF_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_ssim_tests"
-        ;;
-    "training")
-        log "Running training tests..."
-        MODAL_COMMAND="$MODAL_ENV WANDB_API_KEY=$WANDB_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_training_tests"
-        ;;
-    "training_lora")
-        log "Running LoRA training tests..."
-        MODAL_COMMAND="$MODAL_ENV WANDB_API_KEY=$WANDB_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_training_lora_tests"
-        ;;
-    "training_vsa")
-        log "Running training VSA tests..."
-        MODAL_COMMAND="$MODAL_ENV WANDB_API_KEY=$WANDB_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_training_tests_VSA"
-        ;;
-    "inference_sta")
-        log "Running inference STA tests..."
-        MODAL_COMMAND="$MODAL_ENV python3 -m modal run $MODAL_TEST_FILE::run_inference_tests_STA"
-        ;;
-    "kernel_tests")
-        log "Running kernel tests..."
-        MODAL_COMMAND="$MODAL_ENV python3 -m modal run $MODAL_TEST_FILE::run_kernel_tests"
-        ;;
-    "inference_lora")
-        log "Running LoRA tests..."
-        MODAL_COMMAND="$MODAL_ENV python3 -m modal run $MODAL_TEST_FILE::run_inference_lora_tests"
-        ;;
-    "distillation_dmd")
-        log "Running distillation DMD tests..."
-        MODAL_COMMAND="$MODAL_ENV WANDB_API_KEY=$WANDB_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_distill_dmd_tests"
-        ;;
-        # run_inference_tests_vmoba
-    "self_forcing")
-        log "Running self-forcing tests..."
-        MODAL_COMMAND="$MODAL_ENV WANDB_API_KEY=$WANDB_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_self_forcing_tests"
-        ;;
-    "inference_vmoba")
-        log "Running V-MoBA inference tests..."
-        MODAL_COMMAND="$MODAL_ENV python3 -m modal run $MODAL_TEST_FILE::run_inference_tests_vmoba"
-        ;;
-    "unit_test")
-        log "Running unit tests..."
-        MODAL_COMMAND="$MODAL_ENV python3 -m modal run $MODAL_TEST_FILE::run_unit_test"
-        ;;
-    "lora_extraction")
-        log "Running LoRA extraction tests..."
-        MODAL_COMMAND="$MODAL_ENV HF_API_KEY=$HF_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_lora_extraction_tests"
-        ;;
-    *)
-        log "Error: Unknown test type: $TEST_TYPE"
-        exit 1
-        ;;
-esac
-
-log "Executing: $MODAL_COMMAND"
-eval "$MODAL_COMMAND"
-TEST_EXIT_CODE=$?
-
-if [ $TEST_EXIT_CODE -eq 0 ]; then
-    log "Modal test completed successfully"
-else
-    log "Error: Modal test failed with exit code: $TEST_EXIT_CODE"
-fi
-
-log "=== Test execution completed with exit code: $TEST_EXIT_CODE ==="
-exit $TEST_EXIT_CODE

.buildkite/scripts/pre_commit.sh

@@ -1,40 +0,0 @@
-#!/bin/bash
-set -uo pipefail
-
-log() {
-  echo "[$(date '+%Y-%m-%d %H:%M:%S')] $1"
-}
-
-log "=== Starting pre-commit checks ==="
-
-cd "$(dirname "$0")/../.."
-PROJECT_ROOT=$(pwd)
-log "Project root: $PROJECT_ROOT"
-
-if ! python3 -m pre_commit --version &> /dev/null; then
-    log "pre-commit not found, installing..."
-    python3 -m pip install --user pre-commit==4.0.1
-    
-    if ! python3 -m pre_commit --version &> /dev/null; then
-        log "Error: Failed to install pre-commit."
-        exit 1
-    fi
-fi
-
-log "Pre-commit version: $(python3 -m pre_commit --version)"
-
-log "Installing/updating pre-commit hooks..."
-python3 -m pre_commit install --install-hooks
-
-log "Running pre-commit checks on all files..."
-python3 -m pre_commit run --all-files
-PRE_COMMIT_EXIT_CODE=$?
-
-if [ $PRE_COMMIT_EXIT_CODE -eq 0 ]; then
-    log "Pre-commit checks completed successfully"
-else
-    log "Error: Pre-commit checks failed with exit code: $PRE_COMMIT_EXIT_CODE"
-fi
-
-log "=== Pre-commit checks completed with exit code: $PRE_COMMIT_EXIT_CODE ==="
-exit $PRE_COMMIT_EXIT_CODE 

.github/ISSUE_TEMPLATE/1-bug-report.yml

@@ -4,6 +4,14 @@ title: "[Bug] "
 labels: ['Bug']
 
 body:
+- type: textarea
+  attributes:
+    label: Environment
+    description: |
+      Please share your environment with us. You can run the command **python fastvideo/utils/collect_env.py** and copy-paste its output below.
+    placeholder: FastVideo version, platform, python version, cuda version...
+  validations:
+    required: true
 - type: textarea
   attributes:
     label: Describe the bug
@@ -17,13 +25,5 @@ body:
       What command or script did you run? Which **model** are you using?
     placeholder: |
       A placeholder for the command.
-  validations:
-    required: true
-- type: textarea
-  attributes:
-    label: Environment
-    description: |
-      Please share your environment with us. You can run the command **python collect_env.py** and copy-paste its output below.
-    placeholder: FastVideo version, platform, python version, cuda version...
   validations:
     required: true

.github/ISSUE_TEMPLATE/3-RFC.yml

@@ -1,56 +0,0 @@
-name: 💬 Request for comments (RFC).
-description: Ask for feedback on major architectural changes or design choices.
-title: "[RFC]: "
-labels: ["RFC"]
-
-body:
-- type: markdown
-  attributes:
-    value: >
-      #### Please take a look at previous [RFCs](https://github.com/hao-ai-lab/FastVideo/issues?q=label%3ARFC+sort%3Aupdated-desc) for reference.
-- type: textarea
-  attributes:
-    label: Motivation.
-    description: >
-      The motivation of the RFC.
-  validations:
-    required: true
-- type: textarea
-  attributes:
-    label: Proposed Change.
-    description: >
-      The proposed change of the RFC.
-  validations:
-    required: true
-- type: textarea
-  attributes:
-    label: Feedback Period.
-    description: >
-      The feedback period of the RFC. Usually at least one week.
-  validations:
-    required: false
-- type: textarea
-  attributes:
-    label: CC List.
-    description: >
-      The list of people you want to CC.
-  validations:
-    required: false
-- type: textarea
-  attributes:
-    label: Any Other Things.
-    description: >
-      Any other things you would like to mention.
-  validations:
-    required: false
-- type: markdown
-  attributes:
-    value: >
-      Thanks for contributing 🎉!
-- type: checkboxes
-  id: askllm
-  attributes:
-    label: Before submitting a new issue...
-    options:
-      - label: Make sure you already searched for relevant issues.
-        required: true

.github/scripts/runpod_api.py

@@ -160,7 +160,8 @@ def execute_command(pod_id):
     setup_steps = [
         "tar -xzf /tmp/repo.tar.gz --no-same-owner -C /workspace/",
         f"cd /workspace/{repo_name}",
-        "source $HOME/.local/bin/env && source /opt/venv/bin/activate",
+        "source /opt/conda/etc/profile.d/conda.sh",
+        "conda activate fastvideo-dev",
         args.test_command
     ]
     

.github/workflows/build-image.yml

@@ -18,12 +18,6 @@ on:
         required: false
         default: false
         type: boolean
-      python_3_12_cuda_12_9:
-        description: 'Build Python 3.12 image Cuda 12.9'
-        required: false
-        default: false
-        type: boolean
-
 
 permissions:
   contents: read
@@ -55,13 +49,4 @@ jobs:
       python_version: '3.12'
       dockerfile_path: docker/Dockerfile.python3.12
       tag_suffix: py3.12
-    secrets: inherit
-
-  build-python-3-12-cuda-12-9:
-    if: ${{ github.event.inputs.python_3_12_cuda_12_9 == 'true' }}
-    uses: ./.github/workflows/build-image-template.yml
-    with:
-      python_version: '3.12'
-      dockerfile_path: docker/Dockerfile.python3.12.cuda12.9.1
-      tag_suffix: py3.12-cuda12.9.1
     secrets: inherit

.github/workflows/docs.yml

@@ -1,65 +1,82 @@
-name: Deploy Documentation
+# Sample workflow for building and deploying a Hugo site to GitHub Pages
+name: Deploy FastVideo Docs to Pages
 
 on:
+  # Runs on pushes targeting the default branch
   push:
-    branches: [ main ]
+    branches:
+      - main
     paths:
-      - 'docs/**'
-      - 'mkdocs.yml'
-      - 'requirements-mkdocs.txt'
-      - '.github/workflows/docs.yml'
+      - "docs/**/*.md"
+      - "fastvideo/v1/examples/**/*.py"
   pull_request:
-    branches: [ main ]
+    branches: 
+      - main
+    types: [opened, ready_for_review, synchronize, reopened]
     paths:
-      - 'docs/**'
-      - 'mkdocs.yml'
-      - 'requirements-mkdocs.txt'
-      - '.github/workflows/docs.yml'
+      - "docs/**/*.md"
+      - "fastvideo/v1/examples/**/*.py"
 
+  # Allows you to run this workflow manually from the Actions tab
+  workflow_dispatch:
+
+# Sets permissions of the GITHUB_TOKEN to allow deployment to GitHub Pages
 permissions:
   contents: read
   pages: write
   id-token: write
 
+# Allow only one concurrent deployment, skipping runs queued between the run in-progress and latest queued.
+# However, do NOT cancel in-progress runs as we want to allow these production deployments to complete.
 concurrency:
   group: "pages"
   cancel-in-progress: false
 
+# Default to bash
+defaults:
+  run:
+    shell: bash
+
 jobs:
+  pre-commit:
+    uses: ./.github/workflows/pre-commit.yml
+
+  # Build job
   build:
     runs-on: ubuntu-latest
+    needs: pre-commit
     steps:
       - name: Checkout
         uses: actions/checkout@v4
-
-      - name: Setup Python
+      - name: Setup Pages
+        id: pages
+        uses: actions/configure-pages@v5
+      - name: Set up Python
         uses: actions/setup-python@v5
         with:
-          python-version: '3.12'
-
+          python-version: "3.10"
       - name: Install dependencies
         run: |
-          python -m pip install --upgrade pip
-          pip install -r requirements-mkdocs.txt
-
-      - name: Setup Pages
-        uses: actions/configure-pages@v4
-
-      - name: Build documentation
-        run: mkdocs build
-
+          cd docs
+          pip install -r requirements-docs.txt
+      - name: Build docs
+        run: |
+          cd docs
+          make clean
+          make html
       - name: Upload artifact
         uses: actions/upload-pages-artifact@v3
         with:
-          path: ./site
+          path: ./docs/build/html
 
+  # Deployment job
   deploy:
     environment:
       name: github-pages
       url: ${{ steps.deployment.outputs.page_url }}
+    if: ${{ github.event_name == 'push' }}
     runs-on: ubuntu-latest
     needs: build
-    if: github.ref == 'refs/heads/main'
     steps:
       - name: Deploy to GitHub Pages
         id: deployment

.github/workflows/fastvideo-kernel-publish.yml

@@ -1,207 +0,0 @@
-name: Publish FastVideo Kernel to PyPI on Version Change
-
-on:
-  push:
-    branches:
-      - main
-    paths:
-      - "fastvideo-kernel/pyproject.toml"
-  workflow_dispatch:
-
-jobs:
-  check-version-change:
-    runs-on: ubuntu-latest
-    outputs:
-      version-changed: ${{ steps.check-version.outputs.changed }}
-      new-version: ${{ steps.check-version.outputs.new-version }}
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 2
-
-      - name: Check if version changed
-        id: check-version
-        run: |
-          cd fastvideo-kernel
-          # Get current commit's version from pyproject.toml
-          # Use ^ to match start of line to avoid matching minimum-version
-          NEW_VERSION=$(grep -oP '^version\s*=\s*"\K[^"]+' pyproject.toml)
-          echo "New version: $NEW_VERSION"
-          
-          # Get previous version from git history
-          # Note: git show expects path relative to repo root
-          OLD_VERSION=$(git show HEAD~1:fastvideo-kernel/pyproject.toml | grep -oP '^version\s*=\s*"\K[^"]+' || echo "0.0.0")
-          echo "Old version: $OLD_VERSION"
-          
-          if [ "$NEW_VERSION" != "$OLD_VERSION" ]; then
-            echo "Version changed from $OLD_VERSION to $NEW_VERSION"
-            echo "changed=true" >> $GITHUB_OUTPUT
-            echo "new-version=$NEW_VERSION" >> $GITHUB_OUTPUT
-          else
-            echo "Version did not change"
-            echo "changed=false" >> $GITHUB_OUTPUT
-          fi
-
-  build_wheels:
-    name: Build Wheel
-    needs: check-version-change
-    if: ${{ needs.check-version-change.outputs.version-changed == 'true' || github.event_name == 'workflow_dispatch' }}
-    runs-on: ${{ matrix.os }}
-
-    strategy:
-      fail-fast: false
-      matrix:
-          os: [ubuntu-22.04]
-          python-version: ['3.10', '3.11', '3.12']
-          torch-cuda:
-            # - torch-version: '2.5.1'
-            #   cuda-version: '12.4.1'
-            #   torch-cuda-short: 'cu124'
-            # - torch-version: '2.6.0'
-            #   cuda-version: '12.6.3'
-            #   torch-cuda-short: 'cu126'
-            # - torch-version: '2.7.1'
-            #   cuda-version: '12.8.0'
-            #   torch-cuda-short: 'cu128'
-            - torch-version: '2.9.1'
-              cuda-version: '12.8.0'
-              torch-cuda-short: 'cu128'
-
-    steps:
-      - name: Free up disk space
-        run: |
-          echo "Initial disk space:"
-          df -h
-          
-          # Remove large directories
-          sudo rm -rf /usr/share/dotnet
-          sudo rm -rf /usr/local/lib/android
-          sudo rm -rf /opt/ghc
-          sudo rm -rf /usr/local/share/boost
-          sudo rm -rf /usr/share/swift
-          sudo rm -rf /usr/local/lib/node_modules
-          sudo rm -rf /usr/local/share/powershell
-          sudo rm -rf /usr/share/rust
-          sudo rm -rf /usr/local/.ghcup
-          
-          # Remove cached files
-          sudo rm -rf /var/lib/apt/lists/*
-          sudo rm -rf /var/cache/apt/archives/*
-          
-          echo "Disk space after cleanup:"
-          df -h
-
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: ${{ matrix.python-version }}
-
-      - name: Install CUDA ${{ matrix.torch-cuda.cuda-version }}
-        uses: Jimver/cuda-toolkit@v0.2.21
-        id: cuda-toolkit
-        with:
-          cuda: ${{ matrix.torch-cuda.cuda-version }}
-          linux-local-args: '["--toolkit"]'
-          method: 'network'
-
-      - name: Install dependencies (GCC, Clang, CUDA Paths, Git)
-        run: |
-          sudo apt update
-          sudo apt install -y git patchelf gcc-11 g++-11 clang-11
-          sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
-
-          # Allow Git to Access Safe Directory
-          git config --global --add safe.directory /__w/FastVideo/FastVideo
-
-          # Set CUDA environment variables
-          export CUDA_HOME=/usr/local/cuda-${{ matrix.torch-cuda.cuda-version }}
-          export PATH=${CUDA_HOME}/bin:${PATH}
-          export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
-          
-          # Verify installation
-          gcc --version
-          g++ --version
-          clang-11 --version
-          nvcc --version
-
-      - name: Install PyTorch ${{ matrix.torch-cuda.torch-version }}+cu${{ matrix.torch-cuda.cuda-version }}
-        run: |
-          pip install --upgrade pip
-          pip install typing-extensions==4.12.2
-          pip install --no-cache-dir torch==${{ matrix.torch-cuda.torch-version }} --index-url https://download.pytorch.org/whl/${{matrix.torch-cuda.torch-cuda-short}}
-          nvcc --version
-          python --version
-          python -c "import torch; print('PyTorch:', torch.__version__)"
-          python -c "import torch; print('CUDA:', torch.version.cuda)"
-          python -c "from torch.utils import cpp_extension; print (cpp_extension.CUDA_HOME)"
-
-      - name: Build wheel
-        run: |
-          export PYTHONPATH=$GITHUB_WORKSPACE:$PYTHONPATH
-          
-          pip install setuptools ninja packaging wheel triton scikit-build-core cmake build
-          
-          cd fastvideo-kernel
-          git submodule update --init --recursive  # Ensure ThunderKittens submodule is initialized
-          # Release builds are produced on GPU-less runners, so force-enable TK and target Hopper.
-          export TORCH_CUDA_ARCH_LIST="9.0a"
-          export CMAKE_ARGS="${CMAKE_ARGS:-} -DFASTVIDEO_KERNEL_BUILD_TK=ON -DCMAKE_CUDA_ARCHITECTURES=90a"
-          
-          # Build standard wheel (no local version suffix) for PyPI
-          python -m build --wheel --outdir dist
-          
-          # Fix the wheel to be manylinux compliant
-          pip install auditwheel
-          # Target manylinux_2_35 (Ubuntu 22.04 native)
-          auditwheel repair dist/*.whl --plat manylinux_2_35_x86_64 -w fixed_dist
-          # Move fixed wheels back to dist for upload consistency
-          rm dist/*.whl
-          mv fixed_dist/*.whl dist/
-
-      - name: Upload wheel artifact
-        # Only upload if it's the "main" CUDA version we want on PyPI
-        # We upload all to artifacts for inspection/GH releases, but give them distinct artifact names
-        uses: actions/upload-artifact@v4
-        with:
-          name: fastvideo_kernel-py${{ matrix.python-version }}-${{ matrix.torch-cuda.torch-cuda-short }}-torch${{ matrix.torch-cuda.torch-version }}
-          path: fastvideo-kernel/dist/*.whl
-          retention-days: 90
-
-  publish_package:
-    name: Publish package
-    needs: [build_wheels, check-version-change]
-    if: ${{ needs.check-version-change.outputs.version-changed == 'true' || github.event_name == 'workflow_dispatch' }}
-    runs-on: ubuntu-22.04
-    permissions:
-      id-token: write  # Needed for OIDC Trusted Publishing
-
-    steps:
-      - uses: actions/checkout@v4
-
-      - uses: actions/setup-python@v5
-        with:
-          python-version: '3.10'
-
-      - name: Download PyPI wheels
-        uses: actions/download-artifact@v4
-        with:
-          path: fastvideo-kernel/dist/
-          pattern: 'fastvideo_kernel-py*'
-          merge-multiple: true
-
-      - name: Build source distribution
-        run: |
-          pip install build scikit-build-core cmake ninja
-          
-          cd fastvideo-kernel
-          # We don't need full CUDA/Torch to just package the source (sdist)
-          python -m build --sdist --outdir dist
-
-      - name: Publish release distributions to PyPI
-        uses: pypa/gh-action-pypi-publish@release/v1
-        with:
-          packages-dir: fastvideo-kernel/dist/

.github/workflows/matchers/mypy.json

@@ -13,4 +13,4 @@
       ]
     }
   ]
-}
+}

.github/workflows/pr-test.yml

@@ -12,11 +12,13 @@ on:
     paths:
       - "fastvideo/**/*.py"
       - ".github/workflows/pr-test.yml"
-      - "pyproject.toml"
-      - "docker/Dockerfile.python3.12"
-      - "csrc/**"
   workflow_dispatch:
     inputs:
+      custom_image:
+        description: "Custom image from this repository (default: fastvideo-dev:latest)"
+        required: false
+        default: "fastvideo-dev:latest"
+        type: string
       run_encoder_test:
         description: "Run encoder-test"
         required: false
@@ -37,41 +39,10 @@ on:
         required: false
         default: false
         type: boolean
-      run_training_test:
-        description: "Run training-test"
-        required: false
-        default: false
-        type: boolean
-      run_training_test_VSA:
-        description: "Run training-test-VSA"
-        required: false
-        default: false
-        type: boolean
-      run_inference_test_STA:
-        description: "Run inference-test-STA"
-        required: false
-        default: false
-        type: boolean
-      run_precision_test_STA:
-        description: "Run precision-test-STA"
-        required: false
-        default: false
-        type: boolean
-      run_precision_test_VSA:
-        description: "Run precision-test-VSA"
-        required: false
-        default: false
-        type: boolean
-      run_unit_test:
-        description: "Run unit-test"
-        required: false
-        default: false
-        type: boolean
 
 env:
   PYTHONUNBUFFERED: "1"
 
-
 concurrency:
   group: pr-test-${{ github.ref }}
   cancel-in-progress: true
@@ -88,79 +59,24 @@ jobs:
       encoder-test: ${{ steps.filter.outputs.encoder-test }}
       vae-test: ${{ steps.filter.outputs.vae-test }}
       transformer-test: ${{ steps.filter.outputs.transformer-test }}
-      training-test: ${{ steps.filter.outputs.training-test }}
-      training-test-VSA: ${{ steps.filter.outputs.training-test-VSA }}
-      inference-test-STA: ${{ steps.filter.outputs.inference-test-STA }}
-      precision-test-STA: ${{ steps.filter.outputs.precision-test-STA }}
-      precision-test-VSA: ${{ steps.filter.outputs.precision-test-VSA }}
-      unit-test: ${{ steps.filter.outputs.unit-test }}
     steps:
       - uses: actions/checkout@v4
       - uses: dorny/paths-filter@v3
         id: filter
         with:
           filters: |
-            # Define reusable path patterns
-            common-paths: &common-paths
-              - 'pyproject.toml'
-              - 'docker/Dockerfile.python3.10'
-              - 'docker/Dockerfile.python3.11'
-              - 'docker/Dockerfile.python3.12'
-            sta-kernel-paths: &sta-kernel-paths
-              - 'csrc/attn/sliding_tile_attn/**'
-              - 'csrc/attn/sliding_tile_attn/tk/**'
-              - 'csrc/attn/sliding_tile_attn/setup.py'
-              - 'csrc/attn/sliding_tile_attn/config_sta.py'
-              - 'csrc/attn/sliding_tile_attn/st_attn.cpp'
-            vsa-kernel-paths: &vsa-kernel-paths
-              - 'csrc/attn/video_sparse_attn/**'
-              - 'csrc/attn/video_sparse_attn/tk/**'
-              - 'csrc/attn/video_sparse_attn/setup.py'
-              - 'csrc/attn/video_sparse_attn/config_vsa.py'
-              - 'csrc/attn/video_sparse_attn/vsa.cpp'
-            vsa-paths: &vsa-paths
-              - 'fastvideo/**'
-              - *common-paths
-              - *vsa-kernel-paths
-            
-            # Actual tests
             encoder-test:
-              - 'fastvideo/models/encoders/**'
-              - 'fastvideo/models/loader/**'
-              - 'fastvideo/tests/encoders/**'
-              - *common-paths
+              - 'fastvideo/v1/models/encoders/**'
+              - 'fastvideo/v1/models/loaders/**'
+              - 'fastvideo/v1/tests/encoders/**'
             vae-test:
-              - 'fastvideo/models/vaes/**'
-              - 'fastvideo/models/loader/**'
-              - 'fastvideo/tests/vaes/**'
-              - *common-paths
+              - 'fastvideo/v1/models/vaes/**'
+              - 'fastvideo/v1/models/loaders/**'
+              - 'fastvideo/v1/tests/vaes/**'
             transformer-test:
-              - 'fastvideo/models/dits/**'
-              - 'fastvideo/models/loader/**'
-              - 'fastvideo/tests/transformers/**'
-              - 'fastvideo/layers/**'
-              - 'fastvideo/attention/**'
-              - *common-paths
-            training-test:
-              - 'fastvideo/**'
-              - *common-paths
-            training-test-VSA:
-              - 'fastvideo/**'
-              - *common-paths
-              - *vsa-kernel-paths
-            inference-test-STA:
-              - 'fastvideo/**'
-              - *common-paths
-              - *sta-kernel-paths
-            precision-test-STA:
-              - *common-paths
-              - *sta-kernel-paths
-            precision-test-VSA:
-              - *common-paths
-              - *vsa-kernel-paths
-            unit-test:
-              - 'fastvideo/**'
-              - *common-paths
+              - 'fastvideo/v1/models/dits/**'
+              - 'fastvideo/v1/models/loaders/**'
+              - 'fastvideo/v1/tests/transformers/**'
 
   encoder-test:
     needs: change-filter
@@ -173,8 +89,8 @@ jobs:
       gpu_type: "NVIDIA A40"
       gpu_count: 1
       volume_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "uv pip install -e .[test] && pytest ./fastvideo/tests/encoders -s"
+      image: "ghcr.io/${{ github.repository }}/${{ github.event.inputs.custom_image || 'fastvideo-dev:latest' }}"
+      test_command: "pip install -e .[test] && pytest ./fastvideo/v1/tests/encoders -s"
       timeout_minutes: 30
     secrets:
       RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
@@ -191,8 +107,8 @@ jobs:
       gpu_type: "NVIDIA A40"
       gpu_count: 1
       volume_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "uv pip install -e .[test] && pytest ./fastvideo/tests/vaes -s"
+      image: "ghcr.io/${{ github.repository }}/${{ github.event.inputs.custom_image || 'fastvideo-dev:latest' }}"
+      test_command: "pip install -e .[test] && pytest ./fastvideo/v1/tests/vaes -s"
       timeout_minutes: 30
     secrets:
       RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
@@ -209,8 +125,8 @@ jobs:
       gpu_type: "NVIDIA L40S"
       gpu_count: 1
       volume_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "uv pip install -e .[test] && pytest ./fastvideo/tests/transformers -s"
+      image: "ghcr.io/${{ github.repository }}/${{ github.event.inputs.custom_image || 'fastvideo-dev:latest' }}"
+      test_command: "pip install -e .[test] && pytest ./fastvideo/v1/tests/transformers -s"
       timeout_minutes: 30
     secrets:
       RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
@@ -219,7 +135,8 @@ jobs:
   ssim-test:
     needs: change-filter
     if: >-
-      github.event_name != 'workflow_dispatch' || (github.event_name == 'workflow_dispatch' && github.event.inputs.run_ssim_test == 'true')
+      (github.event_name != 'workflow_dispatch' && github.event.pull_request.draft == false) || 
+      (github.event_name == 'workflow_dispatch' && github.event.inputs.run_ssim_test == 'true')
     strategy:
       fail-fast: false
       matrix:
@@ -236,149 +153,14 @@ jobs:
       volume_size: 200
       disk_size: 200
       image: "ghcr.io/${{ github.repository }}/fastvideo-dev:${{ matrix.python-version.tag }}"
-      test_command: "uv pip install -e .[test] && pytest ./fastvideo/tests/ssim -vs"
+      test_command: "pip install -e .[test] && pytest ./fastvideo/v1/tests/ssim -vs"
       timeout_minutes: 60
     secrets:
       RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
       RUNPOD_PRIVATE_KEY: ${{ secrets.RUNPOD_PRIVATE_KEY }}
 
-  training-test:
-    needs: change-filter
-    if: >-
-      (github.event_name != 'workflow_dispatch' && needs.change-filter.outputs.training-test == 'true') || 
-      (github.event_name == 'workflow_dispatch' && github.event.inputs.run_training_test == 'true')
-    uses: ./.github/workflows/runpod-test.yml
-    with:
-      job_id: "training-test"
-      gpu_type: "NVIDIA A40"
-      gpu_count: 4
-      volume_size: 100
-      disk_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "wandb login $WANDB_API_KEY && uv pip install -e .[test] && pytest ./fastvideo/tests/training/Vanilla -srP"
-      timeout_minutes: 30
-    secrets:
-      RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
-      RUNPOD_PRIVATE_KEY: ${{ secrets.RUNPOD_PRIVATE_KEY }}
-      WANDB_API_KEY: ${{ secrets.WANDB_API_KEY }}
-
-  training-test-VSA:
-    needs: change-filter
-    if: >-
-      (github.event_name != 'workflow_dispatch' && needs.change-filter.outputs.training-test-VSA == 'true') || 
-      (github.event_name == 'workflow_dispatch' && github.event.inputs.run_training_test_VSA == 'true')
-    uses: ./.github/workflows/runpod-test.yml
-    with:
-      job_id: "training-test-VSA"
-      gpu_type: "NVIDIA H100 NVL"
-      gpu_count: 2
-      volume_size: 100
-      disk_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "wandb login $WANDB_API_KEY && uv pip install -e .[test] && pytest ./fastvideo/tests/training/VSA -srP"
-      timeout_minutes: 30
-    secrets:
-      RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
-      RUNPOD_PRIVATE_KEY: ${{ secrets.RUNPOD_PRIVATE_KEY }}
-      WANDB_API_KEY: ${{ secrets.WANDB_API_KEY }}
-
-  inference-test-STA:
-    needs: change-filter
-    if: >-
-      (github.event_name != 'workflow_dispatch' && needs.change-filter.outputs.inference-test-STA == 'true') || 
-      (github.event_name == 'workflow_dispatch' && github.event.inputs.run_inference_test_STA == 'true')
-    uses: ./.github/workflows/runpod-test.yml
-    with:
-      job_id: "inference-test-STA"
-      gpu_type: "NVIDIA H100 NVL"
-      gpu_count: 2
-      volume_size: 100
-      disk_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "uv pip install -e .[test] && pytest ./fastvideo/tests/inference/STA -srP"
-      timeout_minutes: 30
-    secrets:
-      RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
-      RUNPOD_PRIVATE_KEY: ${{ secrets.RUNPOD_PRIVATE_KEY }}
-
-  precision-test-STA:
-    needs: change-filter
-    if: >-
-      (github.event_name != 'workflow_dispatch' && needs.change-filter.outputs.precision-test-STA == 'true') || 
-      (github.event_name == 'workflow_dispatch' && github.event.inputs.run_precision_test_STA == 'true')
-    uses: ./.github/workflows/runpod-test.yml
-    with:
-      job_id: "precision-test-STA"
-      gpu_type: "NVIDIA H100 NVL"
-      gpu_count: 1
-      volume_size: 100
-      disk_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "uv pip install -e .[test] && python csrc/attn/tests/test_sta.py"
-      timeout_minutes: 30
-    secrets:
-      RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
-      RUNPOD_PRIVATE_KEY: ${{ secrets.RUNPOD_PRIVATE_KEY }}
-
-  precision-test-VSA:
-    needs: change-filter
-    if: >-
-      (github.event_name != 'workflow_dispatch' && needs.change-filter.outputs.precision-test-VSA == 'true') || 
-      (github.event_name == 'workflow_dispatch' && github.event.inputs.run_precision_test_VSA == 'true')
-    uses: ./.github/workflows/runpod-test.yml
-    with:
-      job_id: "precision-test-VSA"
-      gpu_type: "NVIDIA H100 NVL"
-      gpu_count: 1
-      volume_size: 100
-      disk_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "uv pip install -e .[test] && python csrc/attn/tests/test_vsa.py"
-      timeout_minutes: 30
-    secrets:
-      RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
-      RUNPOD_PRIVATE_KEY: ${{ secrets.RUNPOD_PRIVATE_KEY }}
-
-  unit-test:
-    needs: change-filter
-    if: >-
-      (github.event_name != 'workflow_dispatch' && needs.change-filter.outputs.unit-test == 'true') || 
-      (github.event_name == 'workflow_dispatch' && github.event.inputs.run_unit_test == 'true')
-    uses: ./.github/workflows/runpod-test.yml
-    with:
-      job_id: "unit-test"
-      gpu_type: "NVIDIA L40S"
-      gpu_count: 1
-      volume_size: 100
-      disk_size: 100
-      image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-      test_command: "uv pip install -e .[test] && pytest ./fastvideo/dataset/ -vs && pytest ./fastvideo/workflow/ -vs && pytest ./fastvideo/entrypoints/ -vs"
-      timeout_minutes: 30
-    secrets:
-      RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
-      RUNPOD_PRIVATE_KEY: ${{ secrets.RUNPOD_PRIVATE_KEY }}
-
-  # nightly-test:
-  #   if: >-
-  #     (github.event_name == 'workflow_dispatch' && github.event.inputs.run_nightly_test == 'true')
-  #   uses: ./.github/workflows/runpod-test.yml
-  #   with:
-  #     job_id: "nightly-test"
-  #     gpu_type: "NVIDIA A40"
-  #     gpu_count: 4
-  #     volume_size: 100
-  #     disk_size: 100
-  #     image: "ghcr.io/${{ github.repository }}/fastvideo-dev:py3.12-latest"
-  #     test_command: "wandb login $WANDB_API_KEY && uv pip install -e .[test] && pytest ./fastvideo/tests/nightly/test_e2e_overfit_single_sample.py -vs"
-  #     timeout_minutes: 30
-  #   secrets:
-  #     RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
-  #     RUNPOD_PRIVATE_KEY: ${{ secrets.RUNPOD_PRIVATE_KEY }}
-  #     WANDB_API_KEY: ${{ secrets.WANDB_API_KEY }}
-
   runpod-cleanup:
-    # Add other jobs to this list as you create them
-    needs: [encoder-test, vae-test, transformer-test, ssim-test, training-test, training-test-VSA, inference-test-STA, precision-test-STA, precision-test-VSA] 
+    needs: [encoder-test, vae-test, transformer-test, ssim-test] # Add other jobs to this list as you create them
     if: ${{ always() && ((github.event_name != 'workflow_dispatch' && github.event.pull_request.draft == false) || github.event_name == 'workflow_dispatch') }}
     runs-on: ubuntu-latest
     steps:
@@ -395,7 +177,7 @@ jobs:
 
       - name: Cleanup all RunPod instances
         env:
-          JOB_IDS: '["encoder-test", "vae-test", "transformer-test", "ssim-test-py3.10", "ssim-test-py3.11", "ssim-test-py3.12", "training-test", "training-test-VSA", "inference-test-STA", "precision-test-STA", "precision-test-VSA"]'
+          JOB_IDS: '["encoder-test", "vae-test", "transformer-test", "ssim-test-py3.10", "ssim-test-py3.11", "ssim-test-py3.12"]'
           RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
           GITHUB_RUN_ID: ${{ github.run_id }}
         run: python .github/scripts/runpod_cleanup.py

.github/workflows/pre-commit.yml

@@ -10,7 +10,7 @@ jobs:
     - uses: actions/checkout@v4
     - uses: actions/setup-python@v5
       with:
-        python-version: "3.12"
+        python-version: "3.10"
     - run: echo "::add-matcher::.github/workflows/matchers/actionlint.json"
     - run: echo "::add-matcher::.github/workflows/matchers/mypy.json"
     - uses: pre-commit/action@v3.0.1

.github/workflows/publish-comfyui.yml

@@ -1,28 +0,0 @@
-name: Publish to Comfy registry
-on:
-  workflow_dispatch:
-  push:
-    branches:
-      - main
-      - master
-    paths:
-      - "pyproject.toml"
-
-permissions:
-  issues: write
-
-jobs:
-  publish-node:
-    name: Publish Custom Node to registry
-    runs-on: ubuntu-latest
-    if: ${{ github.repository_owner == 'hao-ai-lab' }}
-    steps:
-      - name: Check out code
-        uses: actions/checkout@v4
-        with:
-          submodules: true
-      - name: Publish Custom Node
-        uses: Comfy-Org/publish-node-action@v1
-        with:
-          ## Add your own personal access token to your Github Repository secrets and reference it here.
-          personal_access_token: ${{ secrets.REGISTRY_ACCESS_TOKEN }}

.github/workflows/runpod-test.yml

@@ -43,8 +43,6 @@ on:
         required: true
       RUNPOD_PRIVATE_KEY:
         required: true
-      WANDB_API_KEY:
-        required: false
 
 jobs:
   run-test:
@@ -57,7 +55,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
-          python-version: "3.12"
+          python-version: "3.10"
 
       - name: Set up SSH key
         run: |
@@ -74,7 +72,6 @@ jobs:
           JOB_ID: ${{ inputs.job_id }}
           RUNPOD_API_KEY: ${{ secrets.RUNPOD_API_KEY }}
           GITHUB_RUN_ID: ${{ github.run_id }}
-          WANDB_API_KEY: ${{ secrets.WANDB_API_KEY }}
         timeout-minutes: ${{ inputs.timeout_minutes }}
         run: >-
           python .github/scripts/runpod_api.py 

.github/workflows/sta-publish.yml

@@ -5,7 +5,7 @@ on:
     branches:
       - main
     paths:
-      - "csrc/attn/sliding_tile_attn/setup.py"
+      - "csrc/sliding_tile_attention/setup.py"
   workflow_dispatch:
 
 jobs:
@@ -23,7 +23,7 @@ jobs:
       - name: Check if version changed
         id: check-version
         run: |
-          cd csrc/attn/sliding_tile_attn
+          cd csrc/sliding_tile_attention
           # Get current commit's version
           NEW_VERSION=$(grep -oP 'VERSION\s*=\s*"\K[^"]+' setup.py)
           echo "New version: $NEW_VERSION"
@@ -136,21 +136,19 @@ jobs:
 
       - name: Build wheel
         run: |
-          export PYTHONPATH=$GITHUB_WORKSPACE:$PYTHONPATH
-          
           # We want setuptools >= 49.6.0 otherwise we can't compile the extension if system CUDA version is 11.7 and pytorch cuda version is 11.6
           # https://github.com/pytorch/pytorch/blob/664058fa83f1d8eede5d66418abff6e20bd76ca8/torch/utils/cpp_extension.py#L810
           # However this still fails so I'm using a newer version of setuptools
           pip install setuptools
           pip install ninja packaging wheel
 
-          cd csrc/attn/sliding_tile_attn  # Move into the correct folder
-          git submodule update --init --recursive # Ensure ThunderKittens submodule is initialized
+          cd csrc/sliding_tile_attention  # Move into the correct folder
+          git submodule update --init --recursive tk # Ensure ThunderKittens submodule is initialized
           python setup.py bdist_wheel --dist-dir=dist
 
       - name: Rename wheel file
         run: |
-          cd csrc/attn/sliding_tile_attn
+          cd csrc/sliding_tile_attention
 
           CUDA_SHORT_VERSION=$(echo ${{ matrix.cuda-version }} | cut -d. -f1,2 | sed 's/\.//g')
           TORCH_SHORT_VERSION=$(echo ${{ matrix.torch-version }} | cut -d. -f1,2)
@@ -165,7 +163,7 @@ jobs:
         uses: actions/upload-artifact@v4
         with:
           name: ${{ env.wheel_name }}
-          path: csrc/attn/sliding_tile_attn/dist/*.whl
+          path: csrc/sliding_tile_attention/dist/*.whl
           retention-days: 90
 
   publish_package:
@@ -231,19 +229,17 @@ jobs:
 
       - name: Build source distribution
         run: |
-          export PYTHONPATH=$GITHUB_WORKSPACE:$PYTHONPATH
-          
           # We want setuptools >= 49.6.0 otherwise we can't compile the extension if system CUDA version is 11.7 and pytorch cuda version is 11.6
           # https://github.com/pytorch/pytorch/blob/664058fa83f1d8eede5d66418abff6e20bd76ca8/torch/utils/cpp_extension.py#L810
           # However this still fails so I'm using a newer version of setuptools
           pip install setuptools
           pip install ninja packaging wheel
 
-          cd csrc/attn/sliding_tile_attn  # Move into the correct folder
-          git submodule update --init --recursive # Ensure ThunderKittens submodule is initialized
+          cd csrc/sliding_tile_attention  # Move into the correct folder
+          git submodule update --init --recursive tk # Ensure ThunderKittens submodule is initialized
           python setup.py sdist --dist-dir=dist
 
       - name: Publish release distributions to PyPI
         uses: pypa/gh-action-pypi-publish@release/v1
         with:
-          packages-dir: csrc/attn/sliding_tile_attn/dist/
+          packages-dir: csrc/sliding_tile_attention/dist/

.github/workflows/test.yml

@@ -28,4 +28,4 @@ jobs:
 
       - name: Run Pytest
         run: |
-          pytest --ignore csrc/attn/test
+          pytest --ignore csrc/sliding_tile_attention/test

.github/workflows/vsa-publish.yml

@@ -1,257 +0,0 @@
-name: Publish Video Sparse Attention Kernel to PyPI on Version Change
-
-on:
-  push:
-    branches:
-      - main
-    paths:
-      - "csrc/attn/video_sparse_attn/setup.py"
-  workflow_dispatch:
-
-jobs:
-  check-version-change:
-    runs-on: ubuntu-latest
-    outputs:
-      version-changed: ${{ steps.check-version.outputs.changed }}
-      new-version: ${{ steps.check-version.outputs.new-version }}
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 2
-
-      - name: Check if version changed
-        id: check-version
-        run: |
-          cd csrc/attn/video_sparse_attn
-          # Get current commit's version
-          NEW_VERSION=$(grep -oP 'VERSION\s*=\s*"\K[^"]+' setup.py)
-          echo "New version: $NEW_VERSION"
-          
-          # Get previous version from git history
-          OLD_VERSION=$(git show HEAD~1:./setup.py | grep -oP 'VERSION\s*=\s*"\K[^"]+' || echo "0.0.0")
-          echo "Old version: $OLD_VERSION"
-          
-          if [ "$NEW_VERSION" != "$OLD_VERSION" ]; then
-            echo "Version changed from $OLD_VERSION to $NEW_VERSION"
-            echo "changed=true" >> $GITHUB_OUTPUT
-            echo "new-version=$NEW_VERSION" >> $GITHUB_OUTPUT
-          else
-            echo "Version did not change"
-            echo "changed=false" >> $GITHUB_OUTPUT
-          fi
-
-  build_wheels:
-    name: Build Wheel
-    needs: check-version-change
-    if: ${{ needs.check-version-change.outputs.version-changed == 'true' || github.event_name == 'workflow_dispatch' }}
-    runs-on: ${{ matrix.os }}
-
-    strategy:
-      fail-fast: false
-      matrix:
-          # Using ubuntu-20.04 instead of 22.04 for more compatibility (glibc). Ideally we'd use the
-          # manylinux docker image, but I haven't figured out how to install CUDA on manylinux.
-          os: [ubuntu-22.04]
-          python-version: ['3.10', '3.11', '3.12', '3.13']
-          # For version reference https://pytorch.org/get-started/previous-versions/
-          torch-cuda:
-            - torch-version: '2.5.1'
-              cuda-version: '12.4.1'
-              torch-cuda-short: 'cu124'
-            - torch-version: '2.6.0'
-              cuda-version: '12.6.3'
-              torch-cuda-short: 'cu126'
-            - torch-version: '2.7.1'
-              cuda-version: '12.8.0'
-              torch-cuda-short: 'cu128'
-
-    steps:
-      - name: Free up disk space
-        run: |
-          echo "Initial disk space:"
-          df -h
-          
-          # Remove large directories
-          sudo rm -rf /usr/share/dotnet
-          sudo rm -rf /usr/local/lib/android
-          sudo rm -rf /opt/ghc
-          sudo rm -rf /usr/local/share/boost
-          sudo rm -rf /usr/share/swift
-          sudo rm -rf /usr/local/lib/node_modules
-          sudo rm -rf /usr/local/share/powershell
-          sudo rm -rf /usr/share/rust
-          sudo rm -rf /usr/local/.ghcup
-          
-          # Remove cached files
-          sudo rm -rf /var/lib/apt/lists/*
-          sudo rm -rf /var/cache/apt/archives/*
-          
-          echo "Disk space after cleanup:"
-          df -h
-
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: ${{ matrix.python-version }}
-
-      - name: Install CUDA ${{ matrix.torch-cuda.cuda-version }}
-        uses: Jimver/cuda-toolkit@v0.2.21
-        id: cuda-toolkit
-        with:
-          cuda: ${{ matrix.torch-cuda.cuda-version }}
-          linux-local-args: '["--toolkit"]'
-          method: 'network'
-
-      - name: Install dependencies (GCC, Clang, CUDA Paths, Git)
-        run: |
-          sudo apt update
-          sudo apt install -y git patchelf gcc-11 g++-11 clang-11
-          sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
-
-          # Allow Git to Access Safe Directory
-          git config --global --add safe.directory /__w/FastVideo/FastVideo
-
-          # Set CUDA environment variables
-          export CUDA_HOME=/usr/local/cuda-${{ matrix.torch-cuda.cuda-version }}
-          export PATH=${CUDA_HOME}/bin:${PATH}
-          export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
-          
-          # Verify installation
-          gcc --version
-          g++ --version
-          clang-11 --version
-          nvcc --version
-
-      - name: Install PyTorch ${{ matrix.torch-cuda.torch-version }}+cu${{ matrix.torch-cuda.cuda-version }}
-        run: |
-          pip install --upgrade pip
-          # With python 3.13 and torch 2.5.1, unless we update typing-extensions, we get error
-          # AttributeError: attribute '__default__' of 'typing.ParamSpec' objects is not writable
-          pip install typing-extensions==4.12.2
-          # We want to figure out the CUDA version to download pytorch
-          # e.g. we can have system CUDA version being 11.7 but if torch==1.12 then we need to download the wheel from cu116
-          # see https://github.com/pytorch/pytorch/blob/main/RELEASE.md#release-compatibility-matrix
-          pip install --no-cache-dir torch==${{ matrix.torch-cuda.torch-version }} --index-url https://download.pytorch.org/whl/${{matrix.torch-cuda.torch-cuda-short}}
-          nvcc --version
-          python --version
-          python -c "import torch; print('PyTorch:', torch.__version__)"
-          python -c "import torch; print('CUDA:', torch.version.cuda)"
-          python -c "from torch.utils import cpp_extension; print (cpp_extension.CUDA_HOME)"
-
-      - name: Build wheel
-        run: |
-          export PYTHONPATH=$GITHUB_WORKSPACE:$PYTHONPATH
-          
-          # We want setuptools >= 49.6.0 otherwise we can't compile the extension if system CUDA version is 11.7 and pytorch cuda version is 11.6
-          # https://github.com/pytorch/pytorch/blob/664058fa83f1d8eede5d66418abff6e20bd76ca8/torch/utils/cpp_extension.py#L810
-          # However this still fails so I'm using a newer version of setuptools
-          pip install setuptools
-          pip install ninja packaging wheel
-
-          cd csrc/attn/video_sparse_attn  # Move into the correct folder
-          git submodule update --init --recursive # Ensure ThunderKittens submodule is initialized
-          python setup.py bdist_wheel --dist-dir=dist
-
-      - name: Rename wheel file
-        run: |
-          cd csrc/attn/video_sparse_attn
-
-          CUDA_SHORT_VERSION=$(echo ${{ matrix.torch-cuda.cuda-version }} | cut -d. -f1,2 | sed 's/\.//g')
-          TORCH_SHORT_VERSION=$(echo ${{ matrix.torch-cuda.torch-version }} | cut -d. -f1,2)
-          # Get the correct version format
-          tmpname=cu${CUDA_SHORT_VERSION}torch${TORCH_SHORT_VERSION}
-          wheel_name=$(ls dist/*whl | xargs -n 1 basename | sed "s/-/+$tmpname-/2")
-          # Rename with version information
-          ls dist/*whl |xargs -I {} mv {} dist/${wheel_name}
-          echo "wheel_name=${wheel_name}" >> $GITHUB_ENV
-
-      - name: Upload wheel artifact
-        uses: actions/upload-artifact@v4
-        with:
-          name: ${{ env.wheel_name }}
-          path: csrc/attn/video_sparse_attn/dist/*.whl
-          retention-days: 90
-
-  publish_package:
-    name: Publish package
-    needs: [build_wheels, check-version-change]
-    if: ${{ needs.check-version-change.outputs.version-changed == 'true' || github.event_name == 'workflow_dispatch' }}
-    runs-on: ubuntu-22.04
-    permissions:
-      id-token: write  # Needed for OIDC Trusted Publishing
-
-    steps:
-      - uses: actions/checkout@v4
-
-      - uses: actions/setup-python@v5
-        with:
-          python-version: '3.10'
-
-      - name: Install CUDA 12.4.1
-        uses: Jimver/cuda-toolkit@v0.2.21
-        id: cuda-toolkit
-        with:
-          cuda: 12.4.1
-          linux-local-args: '["--toolkit"]'
-          method: 'network'
-          sub-packages: '["nvcc"]'
-
-      - name: Install dependencies (GCC, Clang, CUDA Paths, Git)
-        run: |
-          sudo apt update
-          sudo apt install -y git patchelf gcc-11 g++-11 clang-11
-          sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
-
-          # Allow Git to Access Safe Directory
-          git config --global --add safe.directory /__w/FastVideo/FastVideo
-
-          # Set CUDA environment variables
-          export CUDA_HOME=/usr/local/cuda-12.4.1
-          export PATH=${CUDA_HOME}/bin:${PATH}
-          export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
-          
-          # Verify installation
-          gcc --version
-          g++ --version
-          clang-11 --version
-          nvcc --version
-
-      - name: Install PyTorch 2.5.1+cu12.4.1
-        run: |
-          pip install --upgrade pip
-          # With python 3.13 and torch 2.5.1, unless we update typing-extensions, we get error
-          # AttributeError: attribute '__default__' of 'typing.ParamSpec' objects is not writable
-          pip install typing-extensions==4.12.2
-          # We want to figure out the CUDA version to download pytorch
-          # e.g. we can have system CUDA version being 11.7 but if torch==1.12 then we need to download the wheel from cu116
-          # see https://github.com/pytorch/pytorch/blob/main/RELEASE.md#release-compatibility-matrix
-          export TORCH_CUDA_VERSION=124
-          pip install --no-cache-dir torch==2.5.1 --index-url https://download.pytorch.org/whl/cu${TORCH_CUDA_VERSION}
-          nvcc --version
-          python --version
-          python -c "import torch; print('PyTorch:', torch.__version__)"
-          python -c "import torch; print('CUDA:', torch.version.cuda)"
-          python -c "from torch.utils import cpp_extension; print (cpp_extension.CUDA_HOME)"
-
-      - name: Build source distribution
-        run: |
-          export PYTHONPATH=$GITHUB_WORKSPACE:$PYTHONPATH
-          
-          # We want setuptools >= 49.6.0 otherwise we can't compile the extension if system CUDA version is 11.7 and pytorch cuda version is 11.6
-          # https://github.com/pytorch/pytorch/blob/664058fa83f1d8eede5d66418abff6e20bd76ca8/torch/utils/cpp_extension.py#L810
-          # However this still fails so I'm using a newer version of setuptools
-          pip install setuptools
-          pip install ninja packaging wheel
-
-          cd csrc/attn/video_sparse_attn  # Move into the correct folder
-          git submodule update --init --recursive # Ensure ThunderKittens submodule is initialized
-          python setup.py sdist --dist-dir=dist
-
-      - name: Publish release distributions to PyPI
-        uses: pypa/gh-action-pypi-publish@release/v1
-        with:
-          packages-dir: csrc/attn/video_sparse_attn/dist/

.gitignore

@@ -14,15 +14,12 @@ wandb/
 *.pt
 cache_dir/
 wandb/
-venv/
-.venv/
 runs/
 samples/
 *validation/
 data/
 outputs/
 outputs_video
-checkpoints/
 sbatch.sh
 *.out
 env
@@ -30,8 +27,7 @@ env
 **/build/
 **.pyc
 **.txt
-*.log
-weights/
+**.json
 
 # Distribution / packaging
 build/
@@ -41,13 +37,10 @@ dist/
 eggs/
 .eggs/
 
-# MkDocs documentation
-site/
-docs/getting_started/examples/
-docs/inference/examples/
-docs/training/examples/
-docs/distillation/examples/
-!requirements-mkdocs.txt
+# Sphinx documentation
+docs/_build/
+docs/source/getting_started/examples/
+docs/source/inference/examples/
 
 # VSCode
 .vscode/
@@ -63,12 +56,7 @@ docs/distillation/examples/
 *.pkl
 
 # Reference videos
-!fastvideo/tests/ssim/reference_videos/**/*.mp4
+!fastvideo/v1/tests/ssim/reference_videos/**/*.mp4
 
 # Static images
-!docs/assets/images/**/*.png
-!comfyui/assets/**/*.png
-!comfyui/assets/**/*.gif
-
-dmd_t2v_output/
-preprocess_output_text/
+!docs/source/_static/images/**/*.png

.gitmodules

@@ -1,6 +1,3 @@
-[submodule "fastvideo-kernel/include/tk"]
-	path = fastvideo-kernel/include/tk
+[submodule "csrc/sliding_tile_attention/tk"]
+	path = csrc/sliding_tile_attention/tk
 	url = https://github.com/HazyResearch/ThunderKittens.git
-[submodule "fastvideo-kernel/include/cutlass"]
-	path = fastvideo-kernel/include/cutlass
-	url = https://github.com/NVIDIA/cutlass.git

.pre-commit-config.yaml

@@ -3,16 +3,18 @@ default_stages:
   - manual # Run in CI
 exclude: |
   (?x)(
-    fastvideo/third_party/.*|
-    fastvideo-kernel/.*|
+    fastvideo/v1/third_party/.*|
+    csrc/.*|
     assets/.*|
     tests/.*|
     demo/.*|
     predict\.py|
     scripts/.*|
-    prompts/.*|
     fastvideo/data_preprocess/.*|
     fastvideo/dataset/.*|
+    fastvideo/distill/.*|
+    fastvideo/distill\.py|
+    fastvideo/distill_adv\.py|
     fastvideo/models/.*|
     fastvideo/sample/.*|
     fastvideo/train\.py|
@@ -20,7 +22,6 @@ exclude: |
     examples/.*|
     .github/workflows/fastvideo-publish.yml|
     .github/workflows/sta-publish.yml|
-    .github/workflows/vsa-publish.yml|
     .github/workflows/build-image-template.yml|
     docs/source/inference/support_matrix.md
   )
@@ -32,7 +33,7 @@ repos:
     args: [--in-place, --verbose]
     additional_dependencies: [toml] # TODO: Remove when yapf is upgraded
 - repo: https://github.com/astral-sh/ruff-pre-commit
-  rev: v0.11.12
+  rev: v0.11.4
   hooks:
   - id: ruff
     args: [--output-format, github, --fix]
@@ -42,12 +43,12 @@ repos:
   - id: codespell
     additional_dependencies: ['tomli']
     args: ['--toml', 'pyproject.toml']
-# - repo: https://github.com/PyCQA/isort
-#   rev: 6.0.1
-#   hooks:
-#   - id: isort
+- repo: https://github.com/PyCQA/isort
+  rev: 6.0.1
+  hooks:
+  - id: isort
 - repo: https://github.com/jackdewinter/pymarkdown
-  rev: v0.9.30
+  rev: v0.9.29
   hooks:
   - id: pymarkdown
     args: [fix]
@@ -59,7 +60,7 @@ repos:
   rev: v1.15.0
   hooks:
   - id: mypy
-    args: [--python-version, '3.10', --follow-imports, "skip", "--disable-error-code", "union-attr", "--disable-error-code", "override" ]
+    args: [--python-version, '3.10', --follow-imports, "skip", ]
     additional_dependencies: [types-cachetools, types-setuptools, types-PyYAML, types-requests]
 - repo: local
   hooks:
@@ -68,7 +69,7 @@ repos:
     entry: bash
     args:
       - -c
-      - 'git ls-files | grep -v "^fastvideo/tests/ssim/" | grep -v "^fastvideo/tests/inference/lora/L40S_reference_videos/" | grep " " && echo "Filenames should not contain spaces!" && exit 1 || exit 0'
+      - 'git ls-files | grep -v "^fastvideo/v1/tests/ssim/" | grep " " && echo "Filenames should not contain spaces!" && exit 1 || exit 0'
     language: system
     always_run: true
     pass_filenames: false

README.md

@@ -1,47 +1,37 @@
 <div align="center">
-<img src=assets/logos/logo.svg width="30%"/>
+<img src=assets/logo.jpg width="30%"/>
 </div>
 
+**FastVideo is a unified framework for accelerated video generation.**
+
+It features a clean, consistent API that works across popular video models, making it easier for developers to author new models and incorporate system- or kernel-level optimizations.
+With FastVideo's optimizations, you can achieve more than 3x inference improvement compared to other systems.
+
 <p align="center">
-     | <a href="https://hao-ai-lab.github.io/FastVideo"><b>Documentation</b></a> | <a href="https://hao-ai-lab.github.io/FastVideo/inference/inference_quick_start/"><b> Quick Start</b></a> | <a href="https://github.com/hao-ai-lab/FastVideo/discussions/982"  target="_blank"><b>Weekly Dev Meeting</b></a>  | 🟣💬 <a href="https://join.slack.com/t/fastvideo/shared_invite/zt-3f4lao1uq-u~Ipx6Lt4J27AlD2y~IdLQ" target="_blank"> <b>Slack</b> </a> |  🟣💬 <a href="https://ibb.co/c7g1qdD" target="_blank"> <b> WeChat </b> </a> |
+    | <a href="https://hao-ai-lab.github.io/FastVideo"><b>Documentation</b></a> | <a href="https://hao-ai-lab.github.io/FastVideo/inference/inference_quick_start.html"><b> Quick Start</b></a> | 🤗 <a href="https://huggingface.co/FastVideo/FastHunyuan"  target="_blank"><b>FastHunyuan</b></a>  | 🤗 <a href="https://huggingface.co/FastVideo/FastMochi-diffusers" target="_blank"><b>FastMochi</b></a> | 🟣💬 <a href="https://join.slack.com/t/fastvideo/shared_invite/zt-2zf6ru791-sRwI9lPIUJQq1mIeB_yjJg" target="_blank"> <b>Slack</b> </a> |
 </p>
 
-**FastVideo is a unified post-training and inference framework for accelerated video generation.**
-
-## NEWS
-- ```2025/11/19```: Release [CausalWan2.2 I2V A14B Preview](https://huggingface.co/FastVideo/CausalWan2.2-I2V-A14B-Preview-Diffusers) models, [Blog](https://hao-ai-lab.github.io/blogs/fastvideo_causalwan_preview/) and [Inference Code!](https://github.com/hao-ai-lab/FastVideo/blob/main/examples/inference/basic/basic_self_forcing_causal_wan2_2_i2v.py)
-- ```2025/08/04```: Release [FastWan](https://hao-ai-lab.github.io/FastVideo/distillation/dmd) models and [Sparse-Distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/).
-
-<details>
-<summary>More</summary>
-
-- ```2025/06/14```: Release finetuning and inference code for [VSA](https://arxiv.org/pdf/2505.13389)
-- ```2025/04/24```: [FastVideo V1](https://hao-ai-lab.github.io/blogs/fastvideo/) is released!
-- ```2025/02/18```: Release the inference code for [Sliding Tile Attention](https://hao-ai-lab.github.io/blogs/sta/).
-
-</details>
+<div align="center">
+<img src=assets/perf.png width="90%"/>
+</div>
 
 ## Key Features
 
 FastVideo has the following features:
-- End-to-end post-training support for bidirectional and autoregressive models:
-  - Support full finetuning and LoRA finetuning for state-of-the-art open video DiTs
-  - Data preprocessing pipeline for video, image, and text data
-  - Distribution Matching Distillation (DMD2) stepwise distillation.
-  - Sparse attention with [Video Sparse Attention](https://arxiv.org/pdf/2505.13389)
-  - [Sparse distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/) to achineve >50x denoising speedup
-  - Scalable training with FSDP2, sequence parallelism, and selective activation checkpointing.
-  - Causal distillation through Self-Forcing
-  - See this [page](https://hao-ai-lab.github.io/FastVideo/training/overview/) for full list of supported models and recipes.
 - State-of-the-art performance optimizations for inference
-  - Sequence Parallelism for distributed inference
-  - Multiple state-of-the-art attention backends
-  - User-friendly CLI and Python API
-  - See this [page](https://hao-ai-lab.github.io/FastVideo/inference/optimizations/) for full list of supported optimizations.
-- Diverse hardware and OS support
-  - Support H100, A100, 4090
-  - Support Linux, Windows, MacOS
-  - See this [page](https://hao-ai-lab.github.io/FastVideo/inference/hardware_support/) for full list of supported hardware and OS.
+  - [Sliding Tile Attention](https://arxiv.org/pdf/2502.04507)
+  - [TeaCache](https://arxiv.org/pdf/2411.19108)
+  - [Sage Attention](https://arxiv.org/abs/2410.02367)
+- Cutting edge models
+  - Wan2.1 T2V, I2V
+  - HunyuanVideo
+  - FastHunyuan: consistency distilled video diffusion models for 8x inference speedup.
+  - StepVideo T2V
+- Distillation support
+  - Recipes for video DiT, based on [PCM](https://github.com/G-U-N/Phased-Consistency-Model).
+  - Support distilling/finetuning/inferencing state-of-the-art open video DiTs: 1. Mochi 2. Hunyuan.
+- Scalable training with FSDP, sequence parallelism, and selective activation checkpointing, with near linear scaling to 64 GPUs.
+- Memory efficient finetuning with LoRA, precomputed latent, and precomputed text embeddings.
 
 ## Getting Started
 We recommend using an environment manager such as `Conda` to create a clean environment:
@@ -55,33 +45,19 @@ conda activate fastvideo
 pip install fastvideo
 ```
 
-Please see our [docs](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/) for more detailed installation instructions.
-
-## Sparse Distillation
-For our sparse distillation techniques, please see our [distillation docs](https://hao-ai-lab.github.io/FastVideo/distillation/dmd/) and check out our [blog](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/).
-
-See below for recipes and datasets:
-
-|                                            Model                                              |                                               Sparse Distillation                                                 |                                                  Dataset                                                  |
-|:-------------------------------------------------------------------------------------------:  |:---------------------------------------------------------------------------------------------------------------:  |:--------------------------------------------------------------------------------------------------------: |
-| [FastWan2.1-T2V-1.3B](https://huggingface.co/FastVideo/FastWan2.1-T2V-1.3B-Diffusers)         |    [Recipe](https://github.com/hao-ai-lab/FastVideo/tree/main/examples/distill/Wan2.1-T2V/Wan-Syn-Data-480P)      | [FastVideo Synthetic Wan2.1 480P](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x448x832_600k)      |
-| [FastWan2.1-T2V-14B-Preview](https://huggingface.co/FastVideo/FastWan2.1-T2V-14B-Diffusers)   |                                                   Coming soon!                                                    |   [FastVideo Synthetic Wan2.1 720P](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x768x1280_250k)   |
-| [FastWan2.2-TI2V-5B](https://huggingface.co/FastVideo/FastWan2.2-TI2V-5B-Diffusers)           | [Recipe](https://github.com/hao-ai-lab/FastVideo/tree/main/examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free)   | [FastVideo Synthetic Wan2.2 720P](https://huggingface.co/datasets/FastVideo/Wan2.2-Syn-121x704x1280_32k)  |
+Please see our [docs](https://hao-ai-lab.github.io/FastVideo/getting_started/installation.html) for more detailed installation instructions.
 
 ## Inference
 ### Generating Your First Video
-Here's a minimal example to generate a video using the default settings. Make sure VSA kernels are [installed](https://hao-ai-lab.github.io/FastVideo/video_sparse_attention/installation/). Create a file called `example.py` with the following code:
+Here's a minimal example to generate a video using the default settings. Create a file called `example.py` with the following code:
 
 ```python
-import os
 from fastvideo import VideoGenerator
 
 def main():
-    os.environ["FASTVIDEO_ATTENTION_BACKEND"] = "VIDEO_SPARSE_ATTN"
-
     # Create a video generator with a pre-trained model
     generator = VideoGenerator.from_pretrained(
-        "FastVideo/FastWan2.1-T2V-1.3B-Diffusers",
+        "Wan-AI/Wan2.1-T2V-1.3B-Diffusers",
         num_gpus=1,  # Adjust based on your hardware
     )
 
@@ -106,68 +82,68 @@ Run the script with:
 python example.py
 ```
 
-For a more detailed guide, please see our [inference quick start](https://hao-ai-lab.github.io/FastVideo/inference/inference_quick_start/).
+For a more detailed guide, please see our [inference quick start](https://hao-ai-lab.github.io/FastVideo/inference/inference_quick_start.html).
 
 ### Other docs:
 
-- [Design Overview](https://hao-ai-lab.github.io/FastVideo/design/overview/)
-- [Contribution Guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/)
+- [Design Overview](https://hao-ai-lab.github.io/FastVideo/design/overview.html)
+- [Contribution Guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation.html)
 
 ## Distillation and Finetuning
-- [Distillation Guide](https://hao-ai-lab.github.io/FastVideo/distillation/dmd/)
-<!-- - [Finetuning Guide](https://hao-ai-lab.github.io/FastVideo/training/finetune.html) -->
+- [Distillation Guide](https://hao-ai-lab.github.io/FastVideo/training/distillation.html)
+- [Finetuning Guide](https://hao-ai-lab.github.io/FastVideo/training/finetuning.html)
 
-## Awesome work using FastVideo or our research projects
+## 📑 Development Plan
 
-- [SGLang](https://github.com/sgl-project/sglang/tree/main/python/sglang/multimodal_gen): SGLang's diffusion inference functionality is based  on a fork of FastVideo on Sept. 24, 2025. [![Star](https://img.shields.io/github/stars/sgl-project/sglang.svg?style=social&label=Star)](https://github.com/sgl-project/sglang)
-
-- [DanceGRPO](https://github.com/XueZeyue/DanceGRPO): A  unified framework to adapt Group Relative Policy Optimization (GRPO) to visual generation paradigms. Code based on FastVideo. [![Star](https://img.shields.io/github/stars/XueZeyue/DanceGRPO.svg?style=social&label=Star)](https://github.com/XueZeyue/DanceGRPO)
-- [SRPO](https://github.com/Tencent-Hunyuan/SRPO): A method to directly align the full diffusion trajectory with fine-grained human preference. Code based on FastVideo. [![Star](https://img.shields.io/github/stars/Tencent-Hunyuan/SRPO.svg?style=social&label=Star)](https://github.com/Tencent-Hunyuan/SRPO)
-- [DCM](https://github.com/Vchitect/DCM): Dual-expert consistency model for efficient and high-quality video generation. Code based on FastVideo. [![Star](https://img.shields.io/github/stars/Vchitect/DCM.svg?style=social&label=Star)](https://github.com/Vchitect/DCM)
-- [Hunyuan Video 1.5](https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5): A leading lightweight video generation model, where they proposed SSTA based on Sliding Tile Attention. [![Star](https://img.shields.io/github/stars/Tencent-Hunyuan/HunyuanVideo-1.5.svg?style=social&label=Star)](https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5)
-- [Kandinsky-5.0](https://github.com/kandinskylab/kandinsky-5): A family of diffusion models for video & image generation, where their NABLA attention includes a Sliding Tile Attention branch. [![Star](https://img.shields.io/github/stars/kandinskylab/kandinsky-5.svg?style=social&label=Star)](https://github.com/kandinskylab/kandinsky-5)
-- [LongCat Video](https://github.com/meituan-longcat/LongCat-Video): A foundational video generation model with 13.6B parameters with block-sparse attention similar to Video Sparse Attention. [![Star](https://img.shields.io/github/stars/meituan-longcat/LongCat-Video.svg?style=social&label=Star)](https://github.com/meituan-longcat/LongCat-Video)
+<!-- - More distillation methods -->
+  <!-- - [ ] Add Distribution Matching Distillation -->
+- More models support
+  <!-- - [ ] Add CogvideoX model -->
+  - [x] Add StepVideo to V1
+- Optimization features
+  - [x] Teacache in V1
+  - [x] SageAttention in V1
+- Code updates
+  - [x] V1 Configuration API
+  - [ ] Support Training in V1
+  <!-- - [ ] fp8 support -->
+  <!-- - [ ] faster load model and save model support -->
 
 ## 🤝 Contributing
 
-We welcome all contributions. Please check out our guide [here](https://hao-ai-lab.github.io/FastVideo/contributing/overview/).
-See details in [development roadmap](https://github.com/hao-ai-lab/FastVideo/issues/899).
+We welcome all contributions. Please check out our guide [here](https://hao-ai-lab.github.io/FastVideo/developer_guide/overview.html)
+
 ## Acknowledgement
 We learned and reused code from the following projects:
-- [Wan-Video](https://github.com/Wan-Video)
-- [ThunderKittens](https://github.com/HazyResearch/ThunderKittens)
-- [Triton](https://github.com/triton-lang/triton)
-- [DMD2](https://github.com/tianweiy/DMD2)
+- [PCM](https://github.com/G-U-N/Phased-Consistency-Model)
 - [diffusers](https://github.com/huggingface/diffusers)
+- [OpenSoraPlan](https://github.com/PKU-YuanGroup/Open-Sora-Plan)
 - [xDiT](https://github.com/xdit-project/xDiT)
 - [vLLM](https://github.com/vllm-project/vllm)
 - [SGLang](https://github.com/sgl-project/sglang)
 
-We thank [MBZUAI](https://ifm.mbzuai.ac.ae/), [Anyscale](https://www.anyscale.com/), and [GMI Cloud](https://www.gmicloud.ai/) for their support throughout this project.
+We thank MBZUAI and [Anyscale](https://www.anyscale.com/) for their support throughout this project.
 
 ## Citation
-If you find FastVideo useful, please considering citing our work:
+If you use FastVideo for your research, please cite our paper:
 
 ```bibtex
-@software{fastvideo2024,
-  title        = {FastVideo: A Unified Framework for Accelerated Video Generation},
-  author       = {The FastVideo Team},
-  url          = {https://github.com/hao-ai-lab/FastVideo},
-  month        = apr,
-  year         = {2024},
+@misc{zhang2025fastvideogenerationsliding,
+      title={Fast Video Generation with Sliding Tile Attention},
+      author={Peiyuan Zhang and Yongqi Chen and Runlong Su and Hangliang Ding and Ion Stoica and Zhenghong Liu and Hao Zhang},
+      year={2025},
+      eprint={2502.04507},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2502.04507},
 }
-
-@article{zhang2025vsa,
-  title={Vsa: Faster video diffusion with trainable sparse attention},
-  author={Zhang, Peiyuan and Chen, Yongqi and Huang, Haofeng and Lin, Will and Liu, Zhengzhong and Stoica, Ion and Xing, Eric and Zhang, Hao},
-  journal={arXiv preprint arXiv:2505.13389},
-  year={2025}
-}
-
-@article{zhang2025fast,
-  title={Fast video generation with sliding tile attention},
-  author={Zhang, Peiyuan and Chen, Yongqi and Su, Runlong and Ding, Hangliang and Stoica, Ion and Liu, Zhengzhong and Zhang, Hao},
-  journal={arXiv preprint arXiv:2502.04507},
-  year={2025}
+@misc{ding2025efficientvditefficientvideodiffusion,
+      title={Efficient-vDiT: Efficient Video Diffusion Transformers With Attention Tile},
+      author={Hangliang Ding and Dacheng Li and Runlong Su and Peiyuan Zhang and Zhijie Deng and Ion Stoica and Hao Zhang},
+      year={2025},
+      eprint={2502.06155},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2502.06155},
 }
 ```

__init__.py

@@ -1,15 +0,0 @@
-try:
-    from .comfyui.video_generator.nodes import (NODE_CLASS_MAPPINGS,
-                                                NODE_DISPLAY_NAME_MAPPINGS)
-    WEB_DIRECTORY = "./web"
-    __all__ = [
-        'NODE_CLASS_MAPPINGS', 'NODE_DISPLAY_NAME_MAPPINGS', 'WEB_DIRECTORY'
-    ]
-except ImportError:
-    # ComfyUI environment not available, skip comfyui imports
-    NODE_CLASS_MAPPINGS = {}
-    NODE_DISPLAY_NAME_MAPPINGS = {}
-    WEB_DIRECTORY = "./web"
-    __all__ = [
-        'NODE_CLASS_MAPPINGS', 'NODE_DISPLAY_NAME_MAPPINGS', 'WEB_DIRECTORY'
-    ]

assets/full.svg

@@ -1,18 +0,0 @@
-<svg width="252" height="105" viewBox="0 0 252 105" fill="none" xmlns="http://www.w3.org/2000/svg">
-<path d="M89.4843 55.5457H101.361L87.7028 101H74.638L89.4843 55.5457Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M96.0167 1.00057H112.645L118.583 48.273H104.924L103.737 39.7882H79.9827L67.5117 55.5457H85.3273L43.1638 101H28.3174L22.3789 55.5457H33.6621L38.4129 91.3031L58.604 68.2729H44.3515L96.0167 1.00057ZM100.768 13.1217L87.7028 29.4852H103.143L100.768 13.1217Z" fill="#356CFF"/>
-<path d="M37.2252 1.00057L22.3789 48.273H36.0375L40.7884 30.6974L62.6727 30.6974L69.6727 21.0005L43.7576 21.0004L46.7269 11.9096L77.6727 11.9096L86 1.00057L37.2252 1.00057Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M108.488 55.5457L94.2351 101C94.2351 101 105.518 101 120.959 101C136.399 101 144.078 93.0133 148.276 79.788C152.432 68.0157 153.027 55.5457 136.399 55.5457C119.771 55.5457 108.488 55.5457 108.488 55.5457ZM109.081 90.697L116.802 65.8487C116.802 65.8487 120.959 65.8487 132.242 65.8487C143.525 65.8487 137.586 78.5759 135.211 84.0304C133.307 88.4021 127.491 90.697 122.74 90.697C117.989 90.697 109.081 90.697 109.081 90.697Z" fill="#356CFF"/>
-<path d="M173.188 1.00056L168.625 11.9096C168.625 11.9096 149.386 11.9092 142.525 11.9095C135.664 11.9098 136.586 20.3944 141.337 20.3944H159.747C168.654 20.3944 166.961 33.6899 163.904 38.5761C160.467 44.0675 157.371 48.273 148.463 48.273L125.188 48.273L124 37.97L147.87 37.97C153.808 37.97 156.184 29.4852 151.433 29.4852H131.836C120.142 29.4852 125.897 1.00043 141.337 1.00043L173.188 1.00056Z" fill="#356CFF"/>
-<path d="M179.938 1.00056L175.688 11.9096L191.221 11.9096L179.938 48.273H192.409L203.692 11.9096L219.132 11.9095L223.289 1.00043L179.938 1.00056Z" fill="#356CFF"/>
-<path d="M161.341 55.5457H202.845L198.5 65.8487H169.654L167.279 73.7268H188.5L184.749 82.8177H164.31L161.934 90.697H190.251L186.624 101H146.494L161.341 55.5457Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M230.821 54.9391C255.169 54.9391 251.776 67.0602 249.231 77.9692C246.686 88.8783 240.917 101 217.757 101C194.596 101 195.606 88.8783 199.347 77.9692C203.089 67.0602 206.473 54.9391 230.821 54.9391ZM237.948 77.9692C239.984 70.6965 240.917 65.242 228.446 65.242C215.975 65.242 211.818 71.9087 210.037 77.9692C208.255 84.0298 208.255 91.3025 219.538 91.3025C230.821 91.3025 235.911 85.2419 237.948 77.9692Z" fill="#356CFF"/>
-<path d="M173.188 1.00056L168.625 11.9096C168.625 11.9096 149.386 11.9092 142.525 11.9095C135.664 11.9098 136.586 20.3944 141.337 20.3944M173.188 1.00056C173.188 1.00056 156.777 1.00043 141.337 1.00043M173.188 1.00056L141.337 1.00043M141.337 20.3944C146.088 20.3944 150.839 20.3944 159.747 20.3944M141.337 20.3944H159.747M159.747 20.3944C168.654 20.3944 166.961 33.6899 163.904 38.5761C160.467 44.0675 157.371 48.273 148.463 48.273M148.463 48.273C139.556 48.273 125.188 48.273 125.188 48.273M148.463 48.273L125.188 48.273M125.188 48.273L124 37.97M124 37.97C124 37.97 141.931 37.97 147.87 37.97M124 37.97L147.87 37.97M147.87 37.97C153.808 37.97 156.184 29.4852 151.433 29.4852M151.433 29.4852C146.682 29.4852 138.962 29.4852 131.836 29.4852M151.433 29.4852H131.836M131.836 29.4852C120.142 29.4852 125.897 1.00043 141.337 1.00043M37.2252 1.00057L22.3789 48.273H36.0375L40.7884 30.6974L62.6727 30.6974L69.6727 21.0005L43.7576 21.0004L46.7269 11.9096L77.6727 11.9096L86 1.00057L37.2252 1.00057ZM96.0167 1.00057H112.645L118.583 48.273H104.924L103.737 39.7882H79.9827L67.5117 55.5457H85.3273L43.1638 101H28.3174L22.3789 55.5457H33.6621L38.4129 91.3031L58.604 68.2729H44.3515L96.0167 1.00057ZM87.7028 29.4852L100.768 13.1217L103.143 29.4852H87.7028ZM89.4843 55.5457H101.361L87.7028 101H74.638L89.4843 55.5457ZM108.488 55.5457L94.2351 101C94.2351 101 105.518 101 120.959 101C136.399 101 144.078 93.0133 148.276 79.788C152.432 68.0157 153.027 55.5457 136.399 55.5457C119.771 55.5457 108.488 55.5457 108.488 55.5457ZM116.802 65.8487L109.081 90.697C109.081 90.697 117.989 90.697 122.74 90.697C127.491 90.697 133.307 88.4021 135.211 84.0304C137.586 78.5759 143.525 65.8487 132.242 65.8487C120.959 65.8487 116.802 65.8487 116.802 65.8487ZM179.938 1.00056L175.688 11.9096L191.221 11.9096L179.938 48.273H192.409L203.692 11.9096L219.132 11.9095L223.289 1.00043L179.938 1.00056ZM161.341 55.5457H202.845L198.5 65.8487H169.654L167.279 73.7268H188.5L184.749 82.8177H164.31L161.934 90.697H190.251L186.624 101H146.494L161.341 55.5457ZM230.821 54.9391C255.169 54.9391 251.776 67.0602 249.231 77.9692C246.686 88.8783 240.917 101 217.757 101C194.596 101 195.606 88.8783 199.347 77.9692C203.089 67.0602 206.473 54.9391 230.821 54.9391ZM228.446 65.242C240.917 65.242 239.984 70.6965 237.948 77.9692C235.911 85.2419 230.821 91.3025 219.538 91.3025C208.255 91.3025 208.255 84.0298 210.037 77.9692C211.818 71.9087 215.975 65.242 228.446 65.242Z" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M15.2524 55.5451L21.191 100.999L24.7541 100.999L18.8156 55.5451L15.2524 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M8.12646 55.5451L14.065 100.999L15.2527 100.999L9.31417 55.5451L8.12646 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M1 55.5451L6.93853 100.999L7.53239 100.999L1.59385 55.5451L1 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="0.593853"/>
-<path d="M15.2524 48.2724L30.0988 1H33.6619L18.8156 48.2724H15.2524Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M8.12646 48.2724L22.9728 1H24.1605L9.31417 48.2724H8.12646Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M1 48.2724L15.8463 1H16.4402L1.59385 48.2724H1Z" fill="#356CFF" stroke="#356CFF" stroke-width="0.593853"/>
-<path d="M85.3271 55.5457H67.5116L87 12.7363L44.3513 68.2729H58.6038L43.1636 101L85.3271 55.5457Z" fill="#FDC717" stroke="#FDC717" stroke-width="1.18771" stroke-miterlimit="16"/>
-</svg>

assets/icon-simple.svg

@@ -1,6 +0,0 @@
-<svg width="160" height="93" viewBox="0 0 160 93" fill="none" xmlns="http://www.w3.org/2000/svg">
-<path d="M28.8511 91.66L57.6319 1.86368H64.5394L35.7585 91.66H28.8511Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-<path d="M15.0376 91.66L43.8185 1.86368H46.1209L17.3401 91.66H15.0376Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-<path d="M1.22217 91.66L30.003 1.86366H31.1543L2.3734 91.66H1.22217Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.15122"/>
-<path d="M71.4465 1.86483L42.666 91.6599H69.144L78.3538 58.2746H123.251L129.007 39.855H84.1099L89.866 22.5868H152.032L157.788 1.86483H71.4465Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-</svg>

assets/logos/icon_simple.svg

@@ -1,6 +0,0 @@
-<svg width="160" height="93" viewBox="0 0 160 93" fill="none" xmlns="http://www.w3.org/2000/svg">
-<path d="M28.8511 91.66L57.6319 1.86368H64.5394L35.7585 91.66H28.8511Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-<path d="M15.0376 91.66L43.8185 1.86368H46.1209L17.3401 91.66H15.0376Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-<path d="M1.22217 91.66L30.003 1.86366H31.1543L2.3734 91.66H1.22217Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.15122"/>
-<path d="M71.4465 1.86483L42.666 91.6599H69.144L78.3538 58.2746H123.251L129.007 39.855H84.1099L89.866 22.5868H152.032L157.788 1.86483H71.4465Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-</svg>

assets/logos/logo.svg

@@ -1,18 +0,0 @@
-<svg width="252" height="105" viewBox="0 0 252 105" fill="none" xmlns="http://www.w3.org/2000/svg">
-<path d="M89.4843 55.5457H101.361L87.7028 101H74.638L89.4843 55.5457Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M96.0167 1.00057H112.645L118.583 48.273H104.924L103.737 39.7882H79.9827L67.5117 55.5457H85.3273L43.1638 101H28.3174L22.3789 55.5457H33.6621L38.4129 91.3031L58.604 68.2729H44.3515L96.0167 1.00057ZM100.768 13.1217L87.7028 29.4852H103.143L100.768 13.1217Z" fill="#356CFF"/>
-<path d="M37.2252 1.00057L22.3789 48.273H36.0375L40.7884 30.6974L62.6727 30.6974L69.6727 21.0005L43.7576 21.0004L46.7269 11.9096L77.6727 11.9096L86 1.00057L37.2252 1.00057Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M108.488 55.5457L94.2351 101C94.2351 101 105.518 101 120.959 101C136.399 101 144.078 93.0133 148.276 79.788C152.432 68.0157 153.027 55.5457 136.399 55.5457C119.771 55.5457 108.488 55.5457 108.488 55.5457ZM109.081 90.697L116.802 65.8487C116.802 65.8487 120.959 65.8487 132.242 65.8487C143.525 65.8487 137.586 78.5759 135.211 84.0304C133.307 88.4021 127.491 90.697 122.74 90.697C117.989 90.697 109.081 90.697 109.081 90.697Z" fill="#356CFF"/>
-<path d="M173.188 1.00056L168.625 11.9096C168.625 11.9096 149.386 11.9092 142.525 11.9095C135.664 11.9098 136.586 20.3944 141.337 20.3944H159.747C168.654 20.3944 166.961 33.6899 163.904 38.5761C160.467 44.0675 157.371 48.273 148.463 48.273L125.188 48.273L124 37.97L147.87 37.97C153.808 37.97 156.184 29.4852 151.433 29.4852H131.836C120.142 29.4852 125.897 1.00043 141.337 1.00043L173.188 1.00056Z" fill="#356CFF"/>
-<path d="M179.938 1.00056L175.688 11.9096L191.221 11.9096L179.938 48.273H192.409L203.692 11.9096L219.132 11.9095L223.289 1.00043L179.938 1.00056Z" fill="#356CFF"/>
-<path d="M161.341 55.5457H202.845L198.5 65.8487H169.654L167.279 73.7268H188.5L184.749 82.8177H164.31L161.934 90.697H190.251L186.624 101H146.494L161.341 55.5457Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M230.821 54.9391C255.169 54.9391 251.776 67.0602 249.231 77.9692C246.686 88.8783 240.917 101 217.757 101C194.596 101 195.606 88.8783 199.347 77.9692C203.089 67.0602 206.473 54.9391 230.821 54.9391ZM237.948 77.9692C239.984 70.6965 240.917 65.242 228.446 65.242C215.975 65.242 211.818 71.9087 210.037 77.9692C208.255 84.0298 208.255 91.3025 219.538 91.3025C230.821 91.3025 235.911 85.2419 237.948 77.9692Z" fill="#356CFF"/>
-<path d="M173.188 1.00056L168.625 11.9096C168.625 11.9096 149.386 11.9092 142.525 11.9095C135.664 11.9098 136.586 20.3944 141.337 20.3944M173.188 1.00056C173.188 1.00056 156.777 1.00043 141.337 1.00043M173.188 1.00056L141.337 1.00043M141.337 20.3944C146.088 20.3944 150.839 20.3944 159.747 20.3944M141.337 20.3944H159.747M159.747 20.3944C168.654 20.3944 166.961 33.6899 163.904 38.5761C160.467 44.0675 157.371 48.273 148.463 48.273M148.463 48.273C139.556 48.273 125.188 48.273 125.188 48.273M148.463 48.273L125.188 48.273M125.188 48.273L124 37.97M124 37.97C124 37.97 141.931 37.97 147.87 37.97M124 37.97L147.87 37.97M147.87 37.97C153.808 37.97 156.184 29.4852 151.433 29.4852M151.433 29.4852C146.682 29.4852 138.962 29.4852 131.836 29.4852M151.433 29.4852H131.836M131.836 29.4852C120.142 29.4852 125.897 1.00043 141.337 1.00043M37.2252 1.00057L22.3789 48.273H36.0375L40.7884 30.6974L62.6727 30.6974L69.6727 21.0005L43.7576 21.0004L46.7269 11.9096L77.6727 11.9096L86 1.00057L37.2252 1.00057ZM96.0167 1.00057H112.645L118.583 48.273H104.924L103.737 39.7882H79.9827L67.5117 55.5457H85.3273L43.1638 101H28.3174L22.3789 55.5457H33.6621L38.4129 91.3031L58.604 68.2729H44.3515L96.0167 1.00057ZM87.7028 29.4852L100.768 13.1217L103.143 29.4852H87.7028ZM89.4843 55.5457H101.361L87.7028 101H74.638L89.4843 55.5457ZM108.488 55.5457L94.2351 101C94.2351 101 105.518 101 120.959 101C136.399 101 144.078 93.0133 148.276 79.788C152.432 68.0157 153.027 55.5457 136.399 55.5457C119.771 55.5457 108.488 55.5457 108.488 55.5457ZM116.802 65.8487L109.081 90.697C109.081 90.697 117.989 90.697 122.74 90.697C127.491 90.697 133.307 88.4021 135.211 84.0304C137.586 78.5759 143.525 65.8487 132.242 65.8487C120.959 65.8487 116.802 65.8487 116.802 65.8487ZM179.938 1.00056L175.688 11.9096L191.221 11.9096L179.938 48.273H192.409L203.692 11.9096L219.132 11.9095L223.289 1.00043L179.938 1.00056ZM161.341 55.5457H202.845L198.5 65.8487H169.654L167.279 73.7268H188.5L184.749 82.8177H164.31L161.934 90.697H190.251L186.624 101H146.494L161.341 55.5457ZM230.821 54.9391C255.169 54.9391 251.776 67.0602 249.231 77.9692C246.686 88.8783 240.917 101 217.757 101C194.596 101 195.606 88.8783 199.347 77.9692C203.089 67.0602 206.473 54.9391 230.821 54.9391ZM228.446 65.242C240.917 65.242 239.984 70.6965 237.948 77.9692C235.911 85.2419 230.821 91.3025 219.538 91.3025C208.255 91.3025 208.255 84.0298 210.037 77.9692C211.818 71.9087 215.975 65.242 228.446 65.242Z" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M15.2524 55.5451L21.191 100.999L24.7541 100.999L18.8156 55.5451L15.2524 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M8.12646 55.5451L14.065 100.999L15.2527 100.999L9.31417 55.5451L8.12646 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M1 55.5451L6.93853 100.999L7.53239 100.999L1.59385 55.5451L1 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="0.593853"/>
-<path d="M15.2524 48.2724L30.0988 1H33.6619L18.8156 48.2724H15.2524Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M8.12646 48.2724L22.9728 1H24.1605L9.31417 48.2724H8.12646Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M1 48.2724L15.8463 1H16.4402L1.59385 48.2724H1Z" fill="#356CFF" stroke="#356CFF" stroke-width="0.593853"/>
-<path d="M85.3271 55.5457H67.5116L87 12.7363L44.3513 68.2729H58.6038L43.1636 101L85.3271 55.5457Z" fill="#FDC717" stroke="#FDC717" stroke-width="1.18771" stroke-miterlimit="16"/>
-</svg>

benchmarks/fvd/README.md

@@ -1,106 +0,0 @@
-# FVD (Fréchet Video Distance) Benchmark
-
-Evaluate generated video quality using FVD with the I3D feature extractor.
-
-## Quick Start
-
-**Run the benchmark:**
-
-```bash
-bash benchmarks/scripts/run.sh
-```
-
-That's it! The script auto-installs dependencies and runs the benchmark.
-
-**To customize:** Edit `benchmarks/fvd/run_fvd.py` to change:
-- Video paths (`real_dir`, `gen_dir`)
-- Number of videos, frames, sampling strategy
-- Device, batch size, caching, etc.
-
-## Advanced Usage (CLI)
-
-For more control without editing Python files, use the CLI.
-
-**First-time setup** (one-time per pod/environment):
-
-```bash
-bash benchmarks/scripts/setup_fvd.sh
-```
-
-Then run any configuration you want:
-
-```bash
-# Custom configuration
-python -m benchmarks.fvd.cli \
-    --real-path data/real/ \
-    --gen-path outputs/gen/ \
-    --num-videos 1024 \
-    --num-frames 32 \
-    --clip-strategy random \
-    --batch-size 32 \
-    --seed 42 \
-    --extractor clip
-```
-
-**Standard protocols:**
-
-```bash
-# Use predefined protocols
-python -m benchmarks.fvd.cli \
-    --real-path data/real/ \
-    --gen-path outputs/gen/ \
-    --protocol fvd2048_16f  # or fvd2048_128f, quick_test, etc.
-```
-
-This would use i3d model by default as the feature extractor
-
-**Feature caching** (speed up repeated evaluations):
-
-```bash
-python -m benchmarks.fvd.cli \
-    --real-path data/real/ \
-    --gen-path outputs/gen/ \
-    --protocol fvd2048_16f \
-    --cache-real-features fvd-cache/extractor_name  # Directory path (will save/load fvd-cache/extractor_name/extractor-name_real_features.pkl)
-```
-
-Run `python -m benchmarks.fvd.cli --help` for all options.
-
-## Available Protocols
-
-- `fvd2048_16f` - Standard (2048 videos, 16 frames)
-- `fvd2048_128f` - Long videos (128 frames)
-- `fvd2048_128f_subsample8` - Subsampled long videos
-- `quick_test` - Fast testing (10 videos)
-
-## Configuration Options
-
-Key options in `FVDConfig`:
-
-```python
-num_videos=2048,              # Videos to evaluate
-num_frames_per_clip=16,       # Frames per clip
-clip_strategy='beginning',    # beginning|random|uniform|middle|sliding
-frame_stride=1,               # Frame subsampling
-batch_size=32,                # GPU batch size
-device='cuda',                # cuda|cpu
-cache_real_features=None,     # Cache path for speed
-seed=42,                      # Reproducibility
-extractor='i3d',              # i3d|clip|videomae
-```
-
-## Programmatic Usage
-
-```python
-from benchmarks.fvd import compute_fvd_with_config, FVDConfig
-
-config = FVDConfig.fvd2048_16f()  # or custom config
-results = compute_fvd_with_config('data/real/', 'outputs/gen/', config)
-print(f"FVD: {results['fvd']:.2f}")
-```
-
-## Notes
-
-- Requires minimum 10 frames per clip
-- Supports both video files (.mp4, .avi, etc.) and frame directories
-- `--cache-real-features` expects a **directory path** (e.g., `cache/real`), it will automatically create/load `real_features.pkl` inside that directory

benchmarks/fvd/__init__.py

@@ -1,38 +0,0 @@
-"""
-FastVideo Frechet Video Distance (FVD) Benchmark Module.
-    >>> from fastvideo.benchmarks.fvd import compute_fvd_with_config, FVDConfig
-    >>> config = FVDConfig.fvd2048_16f()  # Standard protocol
-    >>> results = compute_fvd_with_config('data/real/', 'outputs/gen/', config)
-    >>> print(f"FVD: {results['fvd']:.2f}")
-"""
-
-from .fvd import (
-    compute_fvd,
-    compute_fvd_with_config,
-    compute_frechet_distance,
-    compute_statistics,
-    FVDConfig,
-)
-from .feature_extractors import (BaseFeatureExtractor, I3DFeatureExtractor,
-                                 load_extractor)
-from .video_utils import (
-    load_video_auto,
-    sample_clips_from_video,
-    load_video_clips_streaming,
-    ClipSamplingStrategy,
-)
-
-__all__ = [
-    'compute_fvd',
-    'compute_fvd_with_config',
-    'compute_frechet_distance',
-    'compute_statistics',
-    'FVDConfig',
-    'BaseFeatureExtractor',
-    'I3DFeatureExtractor',
-    'load_extractor',
-    'load_video_auto',
-    'sample_clips_from_video',
-    'load_video_clips_streaming',
-    'ClipSamplingStrategy',
-]

benchmarks/fvd/cli.py

@@ -1,107 +0,0 @@
-import argparse
-import sys
-import traceback
-from .fvd import compute_fvd_with_config, FVDConfig
-
-
-def main() -> int:
-    parser = argparse.ArgumentParser(
-        description='Compute Fréchet Video Distance (FVD)')
-
-    # Required arguments
-    parser.add_argument('--real-path',
-                        type=str,
-                        required=True,
-                        help='Path to real videos')
-    parser.add_argument('--gen-path',
-                        type=str,
-                        required=True,
-                        help='Path to generated videos')
-
-    # Extractor selection
-    parser.add_argument('--extractor',
-                        type=str,
-                        default='i3d',
-                        choices=['i3d', 'clip', 'videomae'],
-                        help='Feature extractor model to use (default: i3d)')
-
-    # Standard args
-    parser.add_argument('--seed',
-                        type=int,
-                        default=None,
-                        help='Random seed for reproducibility')
-    parser.add_argument('--protocol',
-                        type=str,
-                        default=None,
-                        choices=['fvd2048_16f', 'fvd2048_128f', 'quick_test'],
-                        help='Use standard protocol (overrides other settings)')
-    parser.add_argument('--num-videos',
-                        type=int,
-                        default=2048,
-                        help='Number of videos to use')
-    parser.add_argument('--num-frames',
-                        type=int,
-                        default=16,
-                        help='Number of frames per clip')
-    parser.add_argument('--clip-strategy',
-                        type=str,
-                        default='beginning',
-                        help='Clip sampling strategy')
-    parser.add_argument('--batch-size',
-                        type=int,
-                        default=32,
-                        help='Batch size for feature extraction')
-    parser.add_argument('--device',
-                        type=str,
-                        default='cuda',
-                        help='Device to use (cuda or cpu)')
-    parser.add_argument('--cache-real-features',
-                        type=str,
-                        default=None,
-                        help='Path to cache real video features')
-    parser.add_argument('--quiet',
-                        action='store_true',
-                        help='Suppress progress output')
-
-    args = parser.parse_args()
-
-    # Create config
-    if args.protocol:
-        protocol_map = {
-            'fvd2048_16f': FVDConfig.fvd2048_16f,
-            'fvd2048_128f': FVDConfig.fvd2048_128f,
-            'quick_test': FVDConfig.quick_test,
-        }
-        config = protocol_map[args.protocol]()
-        # Apply overrides
-        config.device = args.device
-        config.cache_real_features = args.cache_real_features
-        config.extractor_model = args.extractor  # Apply extractor arg
-    else:
-        config = FVDConfig(
-            num_videos=args.num_videos,
-            num_frames_per_clip=args.num_frames,
-            extractor_model=args.extractor,  # Apply extractor arg
-            clip_strategy=args.clip_strategy,
-            batch_size=args.batch_size,
-            device=args.device,
-            cache_real_features=args.cache_real_features,
-            seed=args.seed)
-
-    try:
-        _ = compute_fvd_with_config(
-            args.real_path,  # noqa: F841
-            args.gen_path,
-            config,
-            verbose=not args.quiet)
-
-        return 0
-
-    except Exception as e:
-        print(f"Error: {e}", file=sys.stderr)
-        traceback.print_exc(file=sys.stderr)
-        return 1
-
-
-if __name__ == '__main__':
-    sys.exit(main())

benchmarks/fvd/feature_extractors.py

@@ -1,264 +0,0 @@
-"""
-Pluggable Feature Extractors for FVD Computation.
-Supports I3D (standard), CLIP, and VideoMAE via a common interface.
-"""
-
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-from abc import ABC, abstractmethod
-from huggingface_hub import hf_hub_download
-from tqdm import tqdm
-
-try:
-    from transformers import CLIPModel, CLIPProcessor, VideoMAEModel
-    TRANSFORMERS_AVAILABLE = True
-except ImportError:
-    TRANSFORMERS_AVAILABLE = False
-
-
-class BaseFeatureExtractor(ABC, nn.Module):
-    """Abstract base class for all video feature extractors."""
-
-    def __init__(self, device: str = 'cuda'):
-        super().__init__()
-        self.device = torch.device(
-            device if torch.cuda.is_available() else 'cpu')
-
-    @property
-    @abstractmethod
-    def feature_dim(self) -> int:
-        """Dimension of the output feature vector."""
-        pass
-
-    @abstractmethod
-    def preprocess(self, videos: torch.Tensor) -> torch.Tensor:
-        """
-        Args:
-            videos: [B, T, C, H, W] in [0, 255] range.
-        Returns:
-            Preprocessed tensor ready for the model.
-        """
-        pass
-
-    @abstractmethod
-    def extract_features_batch(self, videos: torch.Tensor) -> torch.Tensor:
-        """
-        Extract features for a single batch.
-        Args:
-            videos: [B, T, C, H, W] (raw input)
-        Returns:
-            Features: [B, feature_dim]
-        """
-        pass
-
-    @torch.no_grad()
-    def extract_features(self,
-                         videos: torch.Tensor,
-                         batch_size: int = 32,
-                         verbose: bool = True) -> torch.Tensor:
-        """
-        Extract features for a large tensor of videos by batching.
-        """
-        N = len(videos)
-        all_features = []
-
-        iterator = range(0, N, batch_size)
-        if verbose:
-            iterator = tqdm(
-                iterator,
-                desc=f"Extracting features ({self.__class__.__name__})")
-
-        for i in iterator:
-            batch = videos[i:i + batch_size].to(self.device)
-            features = self.extract_features_batch(batch)
-            all_features.append(features.cpu())
-
-        return torch.cat(all_features, dim=0)
-
-
-# 1. I3D Extractor (The Standard FVD Metric)
-class I3DFeatureExtractor(BaseFeatureExtractor):
-    REPO_ID = 'flateon/FVD-I3D-torchscript'
-    MODEL_FILENAME = 'i3d_torchscript.pt'
-
-    def __init__(self, device: str = 'cuda', cache_dir: str | None = None):
-        super().__init__(device)
-        self.cache_dir = cache_dir
-        self.model = self._load_model()
-        self.model.eval()
-        self.model.to(self.device)
-
-    @property
-    def feature_dim(self) -> int:
-        return 400
-
-    def _load_model(self) -> torch.nn.Module:
-        try:
-            model_path = hf_hub_download(repo_id=self.REPO_ID,
-                                         filename=self.MODEL_FILENAME,
-                                         cache_dir=self.cache_dir)
-            return torch.jit.load(model_path, map_location=self.device)
-        except Exception as e:
-            raise RuntimeError(f"Failed to load I3D model: {e}") from e
-
-    def preprocess(self, videos: torch.Tensor) -> torch.Tensor:
-        """Standard I3D preprocessing: Resize to 224, Norm to [-1, 1]."""
-        B, T, C, H, W = videos.shape
-
-        if T < 10:
-            raise ValueError(f"I3D requires at least 10 frames, got {T}")
-
-        # Normalize to [0, 1]
-        if videos.max() > 1.0:
-            videos = videos / 255.0
-
-        # Scale to [-1, 1]
-        videos = videos * 2.0 - 1.0
-
-        # Resize to 224x224
-        if H != 224 or W != 224:
-            videos = videos.reshape(B * T, C, H, W)
-            videos = F.interpolate(videos,
-                                   size=(224, 224),
-                                   mode='bilinear',
-                                   align_corners=False)
-            videos = videos.reshape(B, T, C, 224, 224)
-
-        # [B, T, C, H, W] -> [B, C, T, H, W]
-        return videos.permute(0, 2, 1, 3, 4).contiguous()
-
-    def extract_features_batch(self, videos: torch.Tensor) -> torch.Tensor:
-        batch = self.preprocess(videos)
-        # TorchScript I3D returns raw logits when return_features=True
-        return self.model(batch,
-                          rescale=False,
-                          resize=False,
-                          return_features=True)
-
-
-# 2. CLIP Extractor (Semantic/Content Quality)
-class CLIPFeatureExtractor(BaseFeatureExtractor):
-
-    def __init__(self,
-                 device: str = 'cuda',
-                 model_name: str = "openai/clip-vit-base-patch32"):
-        if not TRANSFORMERS_AVAILABLE:
-            raise ImportError(
-                "Please install transformers: pip install transformers")
-        super().__init__(device)
-        self.processor = CLIPProcessor.from_pretrained(model_name)
-        self.model = CLIPModel.from_pretrained(model_name).to(self.device)
-        self.model.eval()
-        self._feature_dim = self.model.config.projection_dim
-
-    @property
-    def feature_dim(self) -> int:
-        return self._feature_dim
-
-    def preprocess(self, videos: torch.Tensor) -> torch.Tensor:
-        # Ensure values are [0, 255]
-        if videos.max() <= 1.0:
-            videos = videos * 255.0
-
-        return videos.to(torch.uint8)
-
-    def extract_features_batch(self, videos: torch.Tensor) -> torch.Tensor:
-        # Input: [B, T, C, H, W]
-        B, T, C, H, W = videos.shape
-        videos = self.preprocess(videos)
-
-        # Flatten B*T to treat frames as images
-        images = videos.view(B * T, C, H, W)
-
-        # HF Processor
-        inputs = self.processor(images=images,
-                                return_tensors="pt",
-                                padding=True)
-        inputs = {k: v.to(self.device) for k, v in inputs.items()}
-
-        # Extract features [B*T, Dim]
-        outputs = self.model.get_image_features(**inputs)
-
-        # Reshape [B, T, Dim] and Average Pooling over time
-        outputs = outputs.view(B, T, -1)
-        return outputs.mean(dim=1)
-
-
-# 3. VideoMAE Extractor (Structure/Motion Quality)
-class VideoMAEFeatureExtractor(BaseFeatureExtractor):
-
-    def __init__(self,
-                 device: str = 'cuda',
-                 model_name: str = "MCG-NJU/videomae-base"):
-        if not TRANSFORMERS_AVAILABLE:
-            raise ImportError(
-                "Please install transformers: pip install transformers")
-        super().__init__(device)
-        self.model = VideoMAEModel.from_pretrained(model_name).to(self.device)
-        self.model.eval()
-
-        self.register_buffer(
-            'mean',
-            torch.tensor([0.485, 0.456, 0.406],
-                         device=self.device).view(1, 1, 3, 1, 1))
-        self.register_buffer(
-            'std',
-            torch.tensor([0.229, 0.224, 0.225],
-                         device=self.device).view(1, 1, 3, 1, 1))
-
-    @property
-    def feature_dim(self) -> int:
-        return self.model.config.hidden_size
-
-    def preprocess(self, videos: torch.Tensor) -> torch.Tensor:
-        """
-        Efficient GPU-based preprocessing.
-        Input: [B, T, C, H, W] in range [0, 255]
-        """
-        B, T, C, H, W = videos.shape
-
-        # 1. Resize to 224x224
-        if H != 224 or W != 224:
-            videos = videos.view(B * T, C, H, W)
-            videos = F.interpolate(videos,
-                                   size=(224, 224),
-                                   mode='bilinear',
-                                   align_corners=False)
-            videos = videos.view(B, T, C, 224, 224)
-
-        # 2. Normalize to [0, 1]
-        if videos.dtype != torch.float32:
-            videos = videos.float()
-
-        if videos.max() > 1.0:
-            videos = videos / 255.0
-
-        # 3. Apply ImageNet Mean/Std
-        return (videos - self.mean) / self.std
-
-    def extract_features_batch(self, videos: torch.Tensor) -> torch.Tensor:
-        # Input: [B, T, C, H, W]
-
-        # Fast GPU Preprocessing
-        pixel_values = self.preprocess(videos)
-
-        # Forward pass
-        outputs = self.model(pixel_values)
-
-        # Global Average Pooling of last hidden state [B, T_patches, 768] -> [B, 768]
-        return outputs.last_hidden_state.mean(dim=1)
-
-
-# Factory
-def load_extractor(name: str, device: str = 'cuda') -> BaseFeatureExtractor:
-    name = name.lower()
-    if name == 'i3d':
-        return I3DFeatureExtractor(device)
-    elif name == 'clip':
-        return CLIPFeatureExtractor(device)
-    elif name == 'videomae':
-        return VideoMAEFeatureExtractor(device)
-    else:
-        raise ValueError(
-            f"Unknown extractor: {name}. Options: i3d, clip, videomae")

benchmarks/fvd/fvd.py

@@ -1,405 +0,0 @@
-import numpy as np
-import scipy.linalg
-import torch
-from pathlib import Path
-from collections.abc import Iterator
-import pickle
-from dataclasses import dataclass, field
-from .feature_extractors import BaseFeatureExtractor, load_extractor
-from .video_utils import ClipSamplingStrategy, load_video_clips_streaming
-
-
-def compute_statistics(features: np.ndarray) -> tuple[np.ndarray, np.ndarray]:
-    """Compute mean and covariance."""
-    mu = np.mean(features, axis=0)
-    sigma = np.cov(features, rowvar=False)
-    return mu, sigma
-
-
-def compute_frechet_distance(mu1: np.ndarray,
-                             sigma1: np.ndarray,
-                             mu2: np.ndarray,
-                             sigma2: np.ndarray,
-                             eps: float = 1e-6) -> float:
-    """
-    Compute Fréchet distance between two Gaussians.
-    """
-    sigma1 = sigma1 + eps * np.eye(sigma1.shape[0])
-    sigma2 = sigma2 + eps * np.eye(sigma2.shape[0])
-
-    diff = mu1 - mu2
-    mean_distance = np.sum(diff**2)
-
-    trace_sum = np.trace(sigma1 + sigma2)
-
-    covmean = scipy.linalg.sqrtm(sigma1 @ sigma2)
-
-    if np.iscomplexobj(covmean):
-        if not np.allclose(np.diagonal(covmean).imag, 0, atol=1e-3):
-            print(
-                f"Warning: Imaginary component: {np.max(np.abs(covmean.imag))}")
-        covmean = covmean.real
-
-    trace_product = np.trace(covmean)
-
-    fvd = mean_distance + trace_sum - 2 * trace_product
-
-    return float(fvd)
-
-
-@dataclass
-class FVDConfig:
-    # default configuration for FVD computation:
-
-    # Video selection
-    num_videos: int = 2048
-
-    # Feature Extractor Selection
-    extractor_model: str = 'i3d'  # Options: 'i3d', 'clip', 'videomae'
-
-    # Clip sampling
-    num_frames_per_clip: int = 16
-    num_clips_per_video: int = 1
-    clip_strategy: str | ClipSamplingStrategy = 'beginning'
-
-    # Temporal subsampling
-    frame_stride: int = 1  # 1=no subsampling, 2=every 2nd, 8=every 8th
-    temporal_stride: int = 1  # For sliding window clips
-
-    # Data processing
-    video_extensions: list[str] = field(
-        default_factory=lambda: ['.mp4', '.avi', '.mov', '.mkv'])
-    support_frame_dirs: bool = True
-
-    # Computation
-    batch_size: int = 32
-    device: str = 'cuda'
-
-    use_streaming: bool = True
-    resize_before_extraction: bool = True
-
-    # Caching
-    cache_real_features: str | None = None
-    i3d_model_path: str | None = None
-
-    # Reproducibility
-    seed: int | None = None
-
-    @classmethod
-    def fvd2048_16f(cls) -> 'FVDConfig':
-        """Standard FVD protocol: 2048 videos, 16 frames, beginning clip."""
-        return cls(num_videos=2048,
-                   num_frames_per_clip=16,
-                   clip_strategy='beginning',
-                   use_streaming=True)
-
-    @classmethod
-    def fvd2048_128f(cls) -> 'FVDConfig':
-        """Long video protocol: 2048 videos, 128 frames."""
-        return cls(num_videos=2048,
-                   num_frames_per_clip=128,
-                   clip_strategy='beginning',
-                   use_streaming=True)
-
-    @classmethod
-    def quick_test(cls) -> 'FVDConfig':
-        """Quick test config: 100 videos, 16 frames."""
-        return cls(num_videos=100,
-                   num_frames_per_clip=16,
-                   clip_strategy='beginning')
-
-    def to_dict(self) -> dict:
-        """Export config to dict for logging"""
-        d = self.__dict__.copy()
-        d['clip_strategy'] = str(self.clip_strategy)
-        return d
-
-    def __str__(self) -> str:
-        """Human-readable protocol name"""
-        desc = f"FVD_{self.extractor_model.upper()}_{self.num_videos}_{self.num_frames_per_clip}f"
-        if self.frame_stride > 1:
-            desc += f"_subsample{self.frame_stride}"
-        if self.num_clips_per_video > 1:
-            desc += f"_{self.num_clips_per_video}clips"
-        if self.clip_strategy != 'beginning':
-            desc += f"_{self.clip_strategy}"
-        return desc
-
-
-def extract_features_streaming(video_generator: Iterator[torch.Tensor],
-                               extractor: BaseFeatureExtractor,
-                               batch_size: int = 32,
-                               max_clips: int | None = None,
-                               verbose: bool = True) -> np.ndarray:
-    """
-    Extract features from a video clip generator using streaming.
-    """
-    all_features = []
-    batch = []
-
-    if verbose:
-        print(f"Extracting features with batch_size={batch_size}...")
-
-    with torch.no_grad():
-        for clip_count, clip in enumerate(video_generator):
-            batch.append(clip)
-
-            # Process batch when full
-            if len(batch) == batch_size:
-                batch_tensor = torch.stack(batch).to(extractor.device)
-                features = extractor.extract_features_batch(batch_tensor)
-
-                all_features.append(features.detach().cpu().numpy())
-                batch = []
-
-                if verbose and clip_count % (batch_size * 10) == 0:
-                    print(f"Processed {clip_count} clips...")
-
-            if max_clips is not None and clip_count >= max_clips:
-                break
-
-        # Process remaining clips
-        if len(batch) > 0:
-            batch_tensor = torch.stack(batch).to(extractor.device)
-            features = extractor.extract_features_batch(batch_tensor)
-            all_features.append(features.detach().cpu().numpy())
-
-    if len(all_features) == 0:
-        raise RuntimeError("No features extracted - check video loading")
-
-    features = np.concatenate(all_features, axis=0)
-
-    if verbose:
-        print(f"Extracted {len(features)} feature vectors")
-
-    return features
-
-
-def load_or_compute_features(videos: str | Path | torch.Tensor,
-                             extractor: BaseFeatureExtractor,
-                             config: FVDConfig,
-                             cache_path: str | None = None,
-                             cache_name: str = "real_features") -> np.ndarray:
-    """Load features from cache or compute (with streaming support)"""
-
-    if cache_path is not None:
-        script_dir = Path(__file__).parent
-        cache_dir = script_dir / cache_path
-        cache_file = cache_dir / f"{config.extractor_model}_{cache_name}.pkl"
-
-        if cache_file.exists():
-            print(f"Loading cached features from {cache_file}")
-            with open(cache_file, 'rb') as f:
-                features = pickle.load(f)
-
-            # Validate and limit based on config
-            max_features = config.num_videos * config.num_clips_per_video
-
-            if len(features) < max_features:
-                print(
-                    f"WARNING: Cache has {len(features)} features but need {max_features}"
-                )
-                print("Cached features insufficient - will recompute...")
-            elif len(features) > max_features:
-                print(
-                    f"Using {max_features} features from cache (truncated from {len(features)})"
-                )
-                features = features[:max_features]
-                return features
-            else:
-                print(f"Using all {len(features)} cached features")
-                return features
-
-    print("Computing features from scratch...")
-
-    if isinstance(videos, (str | Path)):
-        target_size = (224, 224) if config.resize_before_extraction else None
-
-        video_generator = load_video_clips_streaming(
-            videos,
-            num_frames=config.num_frames_per_clip,
-            max_videos=config.num_videos,
-            clip_strategy=config.clip_strategy,
-            frame_stride=config.frame_stride,
-            num_clips_per_video=config.num_clips_per_video,
-            video_extensions=config.video_extensions,
-            support_frame_dirs=config.support_frame_dirs,
-            target_size=target_size,
-            verbose=True)
-
-        max_clips = config.num_videos * config.num_clips_per_video
-        features = extract_features_streaming(video_generator,
-                                              extractor,
-                                              batch_size=config.batch_size,
-                                              max_clips=max_clips,
-                                              verbose=True)
-    else:
-        print(f"Extracting features from {len(videos)} video tensors...")
-        features = extractor.extract_features(videos,
-                                              batch_size=config.batch_size,
-                                              verbose=True)
-        features = features.numpy()
-
-    # Validate feature count
-    expected_count = config.num_videos * config.num_clips_per_video
-    if len(features) < expected_count:
-        raise ValueError(
-            f"ERROR: Only extracted {len(features)} features, but need {expected_count}!\n"
-            f"Found fewer videos than expected. Check your video directory.")
-    elif len(features) > expected_count:
-        print(f"Truncating {len(features)} features to {expected_count}")
-        features = features[:expected_count]
-
-    # Cache features if requested
-    if cache_path is not None:
-        script_dir = Path(__file__).parent
-        cache_dir = script_dir / cache_path
-        cache_dir.mkdir(parents=True, exist_ok=True)
-        cache_file = cache_dir / f"{config.extractor_model}_{cache_name}.pkl"
-        print(f"Caching features to {cache_file}")
-        with open(cache_file, 'wb') as f:
-            pickle.dump(features, f)
-
-    return features
-
-
-def compute_fvd_with_config(real_videos: str | Path | torch.Tensor,
-                            gen_videos: str | Path | torch.Tensor,
-                            config: FVDConfig,
-                            verbose: bool = True) -> dict:
-    """
-        Compute FVD using a standardized configuration.
-
-        This is the recommended way to compute FVD for reproducibility.
-
-        Args:
-            real_videos: Path or tensors
-            gen_videos: Path or tensors
-            config: FVDConfig specifying protocol
-            verbose: Print progress
-
-        Returns:
-            results: Dictionary with:
-                - 'fvd': FVD score (float)
-                - 'protocol': Protocol name (str)
-                - 'model': Feature extractor model name (str)
-                - 'config': Configuration dict
-
-        Example:
-            >>> config = FVDConfig.fvd2048_16f()
-            >>> results = compute_fvd_with_config('data/real/', 'outputs/gen/', config)
-            >>> print(f"FVD: {results['fvd']:.2f}")
-    """
-
-    # Seed for reproducibility
-    if config.seed is not None:
-        import random as _rnd
-        _rnd.seed(config.seed)
-        np.random.seed(config.seed)
-        torch.manual_seed(config.seed)
-        if torch.cuda.is_available():
-            torch.cuda.manual_seed_all(config.seed)
-
-    if verbose:
-        print("=" * 70)
-        print(f"Computing FVD with protocol: {config}")
-        print(f"Model: {config.extractor_model.upper()}")
-        print("=" * 70)
-        print("\nConfiguration:")
-        for key, value in config.to_dict().items():
-            print(f"  {key}: {value}")
-        print()
-
-    # Initialize Extractor using Factory
-    if verbose:
-        print(
-            f"\nInitializing {config.extractor_model.upper()} model on {config.device}..."
-        )
-
-    extractor = load_extractor(config.extractor_model, device=config.device)
-
-    # Extract features
-    if verbose:
-        print(f"\n{'='*70}")
-        print("Extracting REAL video features...")
-        print(f"{'='*70}")
-
-    real_features = load_or_compute_features(
-        videos=real_videos,
-        extractor=extractor,
-        config=config,
-        cache_path=config.cache_real_features,
-        cache_name="real_features")
-
-    if verbose:
-        print(f"\n{'='*70}")
-        print("Extracting GENERATED video features...")
-        print(f"{'='*70}")
-
-    gen_features = load_or_compute_features(videos=gen_videos,
-                                            extractor=extractor,
-                                            config=config,
-                                            cache_path=None,
-                                            cache_name="gen_features")
-
-    if verbose:
-        print(f"\nReal videos/clips: {len(real_features)}")
-        print(f"Generated videos/clips: {len(gen_features)}")
-        print(f"\n{'='*70}")
-        print("Computing statistics...")
-        print(f"{'='*70}")
-
-    mu_real, sigma_real = compute_statistics(real_features)
-    mu_gen, sigma_gen = compute_statistics(gen_features)
-
-    if verbose:
-        print(f"\n{'='*70}")
-        print("Computing Fréchet distance...")
-        print(f"{'='*70}")
-
-    fvd = compute_frechet_distance(mu_real, sigma_real, mu_gen, sigma_gen)
-
-    if verbose:
-        print(f"\n{'='*70}")
-        print(f"FVD Score ({config.extractor_model.upper()}): {fvd:.4f}")
-        print(f"Protocol: {config}")
-        print(f"{'='*70}\n")
-
-    results = {
-        'fvd': fvd,
-        'protocol': str(config),
-        'model': config.extractor_model,
-        'config': config.to_dict(),
-    }
-
-    return results
-
-
-def compute_fvd(real_videos: str | Path | torch.Tensor,
-                gen_videos: str | Path | torch.Tensor,
-                num_frames: int = 16,
-                batch_size: int = 32,
-                device: str = 'cuda',
-                num_videos: int | None = 2048,
-                cache_real_features: str | None = None,
-                i3d_model_path: str | None = None,
-                seed: int | None = None,
-                verbose: bool = True) -> float:
-    """
-    Backward compatibility wrapper for computing FVD (defaults to I3D).
-    """
-    num_videos = num_videos if num_videos is not None else 2048
-
-    config = FVDConfig(
-        num_videos=num_videos,
-        num_frames_per_clip=num_frames,
-        extractor_model='i3d',  # Default to I3D
-        batch_size=batch_size,
-        device=device,
-        cache_real_features=cache_real_features,
-        i3d_model_path=i3d_model_path,
-        seed=seed,
-    )
-
-    result = compute_fvd_with_config(real_videos, gen_videos, config, verbose)
-    return result['fvd']

benchmarks/fvd/i3d_model.py

@@ -1,142 +0,0 @@
-"""I3D Feature Extractor for FVD Computation"""
-
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-from pathlib import Path
-from huggingface_hub import hf_hub_download
-from tqdm import tqdm
-from contextlib import suppress
-
-
-class I3DFeatureExtractor(nn.Module):
-    """
-    I3D feature extractor for FVD computation.
-    Extracts 400-dimensional features from videos using I3D model
-    trained on Kinetics-400.
-    """
-
-    REPO_ID = 'flateon/FVD-I3D-torchscript'
-    MODEL_FILENAME = 'i3d_torchscript.pt'
-
-    def __init__(self,
-                 device: str = 'cuda',
-                 cache_dir: str | Path | None = None):
-        super().__init__()
-
-        self.device_str = device
-        if device == 'cuda' and not torch.cuda.is_available():
-            print(
-                "Warning: CUDA requested but not available – falling back to CPU"
-            )
-            self.device = torch.device('cpu')
-        else:
-            self.device = torch.device(device)
-
-        self.cache_dir: str | None
-        if cache_dir is not None:
-            self.cache_dir = str(Path(cache_dir).resolve())
-        else:
-            self.cache_dir = None  # Use HF default cache
-
-        self.model = self._load_model()
-        self.model.eval()
-
-        with suppress(Exception):
-            self.model.to(self.device)
-
-    def _load_model(self) -> torch.nn.Module:
-        """Download and load I3D TorchScript model from Hugging Face Hub."""
-        print(f"Loading I3D model from Hugging Face Hub ({self.REPO_ID})...")
-
-        try:
-            # Download model from Hugging Face Hub
-            model_path = hf_hub_download(repo_id=self.REPO_ID,
-                                         filename=self.MODEL_FILENAME,
-                                         cache_dir=self.cache_dir)
-
-            # Load directly to chosen device
-            model = torch.jit.load(model_path, map_location=self.device)
-            print("I3D model loaded successfully")
-            return model
-
-        except Exception as e:
-            raise RuntimeError(
-                f"Failed to load I3D model from Hugging Face Hub. Error: {e}\n"
-                f"Ensure you have internet connection and huggingface_hub installed:\n"
-                f"pip install huggingface_hub") from e
-
-    def preprocess(self, videos: torch.Tensor) -> torch.Tensor:
-        """
-        Preprocess videos for I3D.
-        
-        Args:
-            videos: [B, T, C, H, W], values in [0, 255]
-        
-        Returns:
-            Preprocessed videos [B, C, T, 224, 224] (normalized and resized)
-        """
-        B, T, C, H, W = videos.shape
-
-        if T < 10:
-            raise ValueError(f"I3D requires at least 10 frames, got {T}")
-
-        # Normalize to [0, 1] if needed
-        if videos.max() > 1.0:
-            videos = videos / 255.0
-
-        # Resize to 224x224 if needed
-        if H != 224 or W != 224:
-            videos = videos.reshape(B * T, C, H, W)
-            videos = F.interpolate(videos,
-                                   size=(224, 224),
-                                   mode='bilinear',
-                                   align_corners=False)
-            videos = videos.reshape(B, T, C, 224, 224)
-
-        # Convert to [B, C, T, H, W] format
-        videos = videos.permute(0, 2, 1, 3, 4).contiguous()
-
-        return videos
-
-    @torch.no_grad()
-    def extract_features(self,
-                         videos: torch.Tensor,
-                         batch_size: int = 32,
-                         verbose: bool = True) -> torch.Tensor:
-        """
-        Extract I3D features
-        
-        Args:
-            videos: [N, T, C, H, W], values in [0, 255]
-            batch_size: Batch size for processing
-            verbose: Show progress bar
-        
-        Returns:
-            Features [N, 400]
-        """
-        N = len(videos)
-        all_features = []
-
-        iterator = range(0, N, batch_size)
-        if verbose:
-            iterator = tqdm(iterator, desc="Extracting I3D features")
-
-        for i in iterator:
-            batch = videos[i:i + batch_size].to(self.device)
-            batch = self.preprocess(batch)  # Now returns [B, C, T, H, W]
-
-            # Use the HF model without rescale/resize (we handle it in preprocess)
-            features = self.model(batch,
-                                  rescale=False,
-                                  resize=False,
-                                  return_features=True)
-
-            all_features.append(features.cpu())
-
-        return torch.cat(all_features, dim=0)
-
-    def __call__(self,
-                 videos: torch.Tensor,
-                 batch_size: int = 32) -> torch.Tensor:
-        return self.extract_features(videos, batch_size=batch_size)

benchmarks/fvd/run_fvd.py

@@ -1,54 +0,0 @@
-import sys
-from pathlib import Path
-
-root_dir = Path(__file__).parent.parent.parent
-sys.path.insert(0, str(root_dir))
-
-from benchmarks.fvd.fvd import FVDConfig, compute_fvd_with_config  # noqa: E402
-
-
-def main() -> None:
-    script_dir = Path(__file__).parent.resolve()
-
-    # Define directories
-    real_dir = "benchmarks/data/real_videos"
-    gen_dir = "benchmarks/data/generated_videos"
-
-    # Compare all 3 models
-    models_to_test = ['i3d', 'clip', 'videomae']
-
-    print(f"\n{'='*60}")
-    print("STARTING COMPARISON BENCHMARK")
-    print(f"{'='*60}")
-
-    for model_name in models_to_test:
-        print(f"\n>>> Running evaluation with {model_name.upper()}...")
-
-        try:
-            cfg = FVDConfig(
-                num_videos=650,
-                num_frames_per_clip=16,
-                extractor_model=model_name,
-                clip_strategy='beginning',
-                device='cuda',
-                seed=42,
-                # Use separate cache folders for each model to avoid conflicts
-                cache_real_features=str(script_dir / f'fvd-cache/{model_name}'),
-            )
-
-            results = compute_fvd_with_config(real_dir,
-                                              gen_dir,
-                                              cfg,
-                                              verbose=False)
-            print(f"FVD: {results['fvd']}\nModel: {results['model']}")
-
-        except Exception as e:
-            print(f"{model_name.upper()} Failed: {e}")
-
-    print(f"\n{'='*60}")
-    print("BENCHMARK COMPLETE")
-    print(f"{'='*60}")
-
-
-if __name__ == '__main__':
-    main()

benchmarks/fvd/validate_fvd.py

@@ -1,97 +0,0 @@
-#!/usr/bin/env python3
-import sys
-from pathlib import Path
-import shutil
-import random
-from fvd import compute_fvd_with_config, FVDConfig
-
-script_path = Path(__file__).resolve()
-fastvideo_root = script_path.parent.parent.parent
-sys.path.insert(0, str(fastvideo_root))
-
-
-def split_videos(video_dir: Path, n_per_subset: int = 128, seed: int = 42):
-    subset_a = video_dir.parent / 'bair_full_subset_A'
-    subset_b = video_dir.parent / 'bair_full_subset_B'
-
-    if subset_a.exists():
-        shutil.rmtree(subset_a)
-    if subset_b.exists():
-        shutil.rmtree(subset_b)
-
-    subset_a.mkdir(parents=True)
-    subset_b.mkdir(parents=True)
-
-    videos = sorted(video_dir.glob('*.mp4'))
-
-    random.seed(seed)
-    shuffled = list(videos)
-    random.shuffle(shuffled)
-
-    needed = n_per_subset * 2
-    if len(shuffled) > needed:
-        shuffled = shuffled[:needed]
-
-    mid = len(shuffled) // 2
-
-    print(f"\nSplitting {len(shuffled)} BAIR FULL videos:")
-    print(f"  Subset A: {mid} videos")
-    print(f"  Subset B: {len(shuffled) - mid} videos")
-
-    for v in shuffled[:mid]:
-        shutil.copy2(v, subset_a / v.name)
-
-    for v in shuffled[mid:]:
-        shutil.copy2(v, subset_b / v.name)
-
-    return subset_a, subset_b, mid
-
-
-def validate_fvd(subset_a: Path, subset_b: Path, num_videos: int):
-    config = FVDConfig(num_videos=num_videos,
-                       num_frames_per_clip=16,
-                       clip_strategy='beginning',
-                       batch_size=8,
-                       device='cuda',
-                       seed=42)
-
-    print("\n" + "=" * 70)
-    print("TEST 1: Identity Test")
-    print("=" * 70)
-
-    result1 = compute_fvd_with_config(real_videos=str(subset_a),
-                                      gen_videos=str(subset_a),
-                                      config=config,
-                                      verbose=False)
-    fvd_identity = result1['fvd']
-    print(f"\nIdentity FVD: {fvd_identity:.2f}")
-
-    print("\n" + "=" * 70)
-    print("TEST 2: Real vs Real")
-    print("=" * 70)
-
-    result2 = compute_fvd_with_config(real_videos=str(subset_a),
-                                      gen_videos=str(subset_b),
-                                      config=config,
-                                      verbose=False)
-    fvd_real = result2['fvd']
-    print(f"\nReal vs Real FVD: {fvd_real:.2f}")
-
-    print("\n" + "=" * 70)
-    print("RESULTS")
-    print("=" * 70)
-    print(f"Identity:        {fvd_identity:.2f}")
-    print(f"Real vs Real:    {fvd_real:.2f}")
-
-
-def main() -> None:
-    bair_dir = Path('benchmarks/data/bair_full_videos')
-
-    subset_a, subset_b, count = split_videos(bair_dir,
-                                             n_per_subset=128,
-                                             seed=42)
-    validate_fvd(subset_a, subset_b, count)
-
-
-if __name__ == '__main__':
-    main()

benchmarks/fvd/video_utils.py

@@ -1,490 +0,0 @@
-import torch
-import cv2
-import numpy as np
-from pathlib import Path
-from collections.abc import Iterator
-from tqdm import tqdm
-from enum import Enum
-
-
-class ClipSamplingStrategy(Enum):
-    """Clip sampling strategies for FVD evaluation."""
-    BEGINNING = 'beginning'  # Take first N frames (most common)
-    RANDOM = 'random'  # Random N consecutive frames
-    UNIFORM = 'uniform'  # Uniformly spaced frames across video
-    MIDDLE = 'middle'  # Middle N frames
-    SLIDING = 'sliding'  # Multiple sliding windows
-    ALL = 'all'  # All possible clips
-
-
-def _load_video_cv2(video_path: str | Path,
-                    num_frames: int | None = 16,
-                    sample_strategy: str = 'uniform') -> torch.Tensor:
-    """
-    Load video from video file using OpenCV.
-    
-    Args:
-        video_path: Path to video file (MP4, AVI, MOV, MKV)
-        num_frames: Number of frames to extract
-        sample_strategy: 'uniform' or 'random'
-    
-    Returns:
-        video: [T, C, H, W]
-    """
-    video_path = str(video_path)
-    cap = cv2.VideoCapture(video_path)
-
-    if not cap.isOpened():
-        raise RuntimeError(f"Cannot open video: {video_path}")
-
-    frames = []
-    total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
-
-    if num_frames is None:
-        # Read all available frames
-        while True:
-            ret, frame = cap.read()
-            if not ret:
-                break
-            frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-            frames.append(frame)
-
-        cap.release()
-        if len(frames) == 0:
-            raise RuntimeError(f"Video has 0 frames: {video_path}")
-
-        frames = np.stack(frames)  # [T, H, W, C]
-        frames = torch.from_numpy(frames).permute(0, 3, 1,
-                                                  2).float()  # [T, C, H, W]
-        return frames
-
-    if total_frames == 0:
-        raise RuntimeError(f"Video has 0 frames: {video_path}")
-
-    # Determine frame indices for sampling
-    if total_frames < num_frames:
-        frame_indices = list(range(
-            total_frames)) + [total_frames - 1] * (num_frames - total_frames)
-    elif sample_strategy == 'uniform':
-        frame_indices = np.linspace(0, total_frames - 1, num_frames,
-                                    dtype=int).tolist()
-    elif sample_strategy == 'random':
-        frame_indices = sorted(
-            np.random.choice(total_frames, num_frames, replace=False))
-    else:
-        raise ValueError(f"Unknown sample_strategy: {sample_strategy}")
-
-    # Extract frames
-    for idx in frame_indices:
-        cap.set(cv2.CAP_PROP_POS_FRAMES, idx)
-        ret, frame = cap.read()
-
-        if not ret:
-            if len(frames) > 0:
-                frames.append(frames[-1].copy())
-            else:
-                h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
-                w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
-                frames.append(np.zeros((h, w, 3), dtype=np.uint8))
-            continue
-
-        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-        frames.append(frame)
-
-    cap.release()
-
-    frames = np.stack(frames)  # [T, H, W, C]
-    frames = torch.from_numpy(frames).permute(0, 3, 1,
-                                              2).float()  # [T, C, H, W]
-
-    return frames
-
-
-def _load_video_from_frames(
-        frame_dir: str | Path,
-        num_frames: int | None = 16,
-        sample_strategy: str = 'uniform',
-        frame_extensions: list[str] | None = None) -> torch.Tensor:
-    """
-    Load video from directory of frame images.
-    
-    Args:
-        frame_dir: Directory containing frames
-        num_frames: Number of frames to sample
-        sample_strategy: 'uniform' or 'random'
-        frame_extensions: Image file extensions to look for
-    
-    Returns:
-        video: [T, C, H, W]
-    """
-    if frame_extensions is None:
-        frame_extensions = ['.jpg', '.png', '.jpeg', '.bmp']
-
-    frame_dir = Path(frame_dir)
-
-    if not frame_dir.exists():
-        raise FileNotFoundError(f"Frame directory not found: {frame_dir}")
-
-    # Find all frames
-    frame_files: list[Path] = []
-    for ext in frame_extensions:
-        frame_files.extend(frame_dir.glob(f"*{ext}"))
-
-    if len(frame_files) == 0:
-        raise ValueError(
-            f"No frames found in {frame_dir} with extensions {frame_extensions}"
-        )
-
-    frame_files = sorted(frame_files, key=lambda x: x.name)
-    total_frames = len(frame_files)
-
-    # Determine frame indices
-    if num_frames is None:
-        frame_indices = list(range(total_frames))
-    else:
-        if total_frames < num_frames:
-            frame_indices = list(range(total_frames)) + [total_frames - 1] * (
-                num_frames - total_frames)
-        elif sample_strategy == 'uniform':
-            frame_indices = np.linspace(0,
-                                        total_frames - 1,
-                                        num_frames,
-                                        dtype=int).tolist()
-        elif sample_strategy == 'random':
-            frame_indices = sorted(
-                np.random.choice(total_frames, num_frames, replace=False))
-        else:
-            raise ValueError(f"Unknown sample_strategy: {sample_strategy}")
-
-    # Load frames
-    frames = []
-    for idx in frame_indices:
-        frame_path = frame_files[idx]
-        frame = cv2.imread(str(frame_path))
-
-        if frame is None:
-            raise RuntimeError(f"Failed to load frame: {frame_path}")
-
-        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-        frames.append(frame)
-
-    # Stack and convert to tensor
-    frames = np.stack(frames)  # [T, H, W, C]
-    frames = torch.from_numpy(frames).permute(0, 3, 1,
-                                              2).float()  # [T, C, H, W]
-
-    return frames
-
-
-def _detect_video_format(path: str | Path) -> str:
-    """
-    Detect if path is a video file or frame directory.
-    
-    Returns:
-        'video_file', 'frame_directory', or 'unknown'
-    """
-    path = Path(path)
-
-    if path.is_file():
-        return 'video_file'
-    elif path.is_dir():
-        # Check if contains image files
-        image_extensions = ['.jpg', '.jpeg', '.png', '.bmp']
-        for ext in image_extensions:
-            if list(path.glob(f"*{ext}")):
-                return 'frame_directory'
-        return 'unknown'
-    else:
-        raise ValueError(f"Path does not exist: {path}")
-
-
-def load_video_auto(video_path: str | Path,
-                    num_frames: int | None = 16,
-                    sample_strategy: str = 'uniform') -> torch.Tensor:
-    """
-    Automatically detect format and load video.
-    
-    Supports:
-    - Video files (MP4, AVI, MOV, MKV)
-    - Frame directories (JPG, PNG)
-    
-    Args:
-        video_path: Path to video file or frame directory
-        num_frames: Number of frames to extract
-        sample_strategy: 'uniform' or 'random'
-    
-    Returns:
-        video: [T, C, H, W]
-    """
-    format_type = _detect_video_format(video_path)
-
-    if format_type == 'video_file':
-        return _load_video_cv2(video_path, num_frames, sample_strategy)
-    elif format_type == 'frame_directory':
-        return _load_video_from_frames(video_path, num_frames, sample_strategy)
-    else:
-        raise ValueError(f"Unknown video format at {video_path}")
-
-
-def sample_clips_from_video(
-        video: torch.Tensor,
-        num_frames_per_clip: int = 16,
-        num_clips: int = 1,
-        strategy: str | ClipSamplingStrategy = ClipSamplingStrategy.BEGINNING,
-        frame_stride: int = 1,
-        temporal_stride: int = 1) -> list[torch.Tensor]:
-    """
-    Sample clips from a video with various strategies.
-        
-    Args:
-        video: [T, C, H, W] full video
-        num_frames_per_clip: Frames per clip
-        num_clips: Number of clips to extract
-        strategy: ClipSamplingStrategy or string ('beginning', 'random', etc.)
-        frame_stride: Skip frames (FPS control: 1=all, 2=every 2nd, 8=every 8th)
-        temporal_stride: Stride between clips for sliding window
-    
-    Returns:
-        List of clips, each [num_frames_per_clip, C, H, W]
-    
-    Examples:
-        >>> # Beginning clip (most common for FVD)
-        >>> clips = sample_clips_from_video(video, 16, strategy='beginning')
-        
-        >>> # Multiple random clips
-        >>> clips = sample_clips_from_video(video, 16, num_clips=4, strategy='random')
-        
-        >>> # Subsample FPS by 2x (every 2nd frame)
-        >>> clips = sample_clips_from_video(video, 16, frame_stride=2)
-        
-        >>> # Sliding window with overlap
-        >>> clips = sample_clips_from_video(video, 16, strategy='sliding', temporal_stride=8)
-    """
-    # Convert string to enum if needed
-    if isinstance(strategy, str):
-        strategy = ClipSamplingStrategy(strategy)
-
-    T, C, H, W = video.shape
-
-    # Apply frame stride (FPS subsampling)
-    if frame_stride > 1:
-        video = video[::frame_stride]
-        T = len(video)
-
-    effective_clip_length = num_frames_per_clip
-
-    # Handle videos shorter than clip length
-    if effective_clip_length > T:
-        pad_length = effective_clip_length - T
-        last_frame = video[-1:].repeat(pad_length, 1, 1, 1)
-        video = torch.cat([video, last_frame], dim=0)
-        T = len(video)
-
-    clips = []
-
-    if strategy == ClipSamplingStrategy.BEGINNING:
-        # Take first clip (most common for FVD evaluation)
-        clip = video[:effective_clip_length]
-        clips.append(clip)
-
-    elif strategy == ClipSamplingStrategy.MIDDLE:
-        # Take middle clip
-        start = (T - effective_clip_length) // 2
-        clip = video[start:start + effective_clip_length]
-        clips.append(clip)
-
-    elif strategy == ClipSamplingStrategy.RANDOM:
-        # Sample N random clips
-        for _ in range(num_clips):
-            if effective_clip_length == T:
-                start = 0
-            else:
-                start = np.random.randint(0, T - effective_clip_length + 1)
-            clip = video[start:start + effective_clip_length]
-            clips.append(clip)
-
-    elif strategy == ClipSamplingStrategy.UNIFORM:
-        # Uniformly spaced clips
-        if num_clips == 1:
-            # Single clip from middle
-            start = (T - effective_clip_length) // 2
-            clip = video[start:start + effective_clip_length]
-            clips.append(clip)
-        else:
-            # Multiple uniformly spaced clips
-            step = (T - effective_clip_length) / (num_clips -
-                                                  1) if num_clips > 1 else 0
-            for i in range(num_clips):
-                start = int(i * step)
-                start = min(start, T - effective_clip_length)
-                clip = video[start:start + effective_clip_length]
-                clips.append(clip)
-
-    elif strategy == ClipSamplingStrategy.SLIDING:
-        # Sliding window with stride
-        for start in range(0, T - effective_clip_length + 1, temporal_stride):
-            clip = video[start:start + effective_clip_length]
-            clips.append(clip)
-            if len(clips) >= num_clips:
-                break
-
-    elif strategy == ClipSamplingStrategy.ALL:
-        # All possible clips (overlapping)
-        for start in range(T - effective_clip_length + 1):
-            clip = video[start:start + effective_clip_length]
-            clips.append(clip)
-
-    else:
-        raise ValueError(f"Unknown strategy: {strategy}")
-
-    return clips
-
-
-def load_video_clips_streaming(directory: str | Path,
-                               num_frames: int = 16,
-                               max_videos: int | None = None,
-                               clip_strategy: str
-                               | ClipSamplingStrategy = 'beginning',
-                               frame_stride: int = 1,
-                               num_clips_per_video: int = 1,
-                               video_extensions: list[str] | None = None,
-                               support_frame_dirs: bool = True,
-                               target_size: tuple[int, int] | None = (224, 224),
-                               verbose: bool = True) -> Iterator[torch.Tensor]:
-    """
-    This generator yields clips one-by-one instead of loading all videos into RAM.
-    Perfect for large datasets where memory is limited.
-    
-    Args:
-        directory: Path to directory with videos
-        num_frames: Frames per clip
-        max_videos: Max videos to load
-        clip_strategy: 'beginning', 'random', 'uniform', etc.
-        frame_stride: Frame skip (1=all, 2=every 2nd, 8=every 8th)
-        num_clips_per_video: Number of clips per video
-        video_extensions: Video file extensions
-        support_frame_dirs: Also load frame directories
-        target_size: Resize clips to (H, W). If None, keep original size.
-        verbose: Show progress
-    
-    Yields:
-        clip: [T, C, H, W] individual clips
-    
-    Example:
-        >>> for clip in load_video_clips_streaming('data/videos/', num_frames=16):
-        >>>     features = model.extract_features(clip.unsqueeze(0))
-        >>>     # Process one clip at a time - low memory usage!
-    """
-    if video_extensions is None:
-        video_extensions = ['.mp4', '.avi', '.mov', '.mkv']
-
-    directory = Path(directory)
-
-    if not directory.exists():
-        raise FileNotFoundError(f"Directory not found: {directory}")
-
-    # Find video paths
-    video_paths: list[Path] = []
-
-    # Find video files
-    for ext in video_extensions:
-        video_paths.extend(directory.glob(f"**/*{ext}"))
-
-    # Find frame directories if enabled
-    if support_frame_dirs:
-        for subdir in directory.iterdir():
-            if subdir.is_dir():
-                # Check if it contains frames
-                image_extensions = ['.jpg', '.jpeg', '.png', '.bmp']
-                for ext in image_extensions:
-                    if list(subdir.glob(f"*{ext}")):
-                        video_paths.append(subdir)
-                        break
-
-    if len(video_paths) == 0:
-        raise ValueError(f"No videos found in {directory}")
-
-    video_paths = sorted(video_paths)
-
-    if max_videos is not None:
-        video_paths = video_paths[:max_videos]
-
-    if verbose:
-        print(f"Found {len(video_paths)} videos in {directory}")
-        if num_clips_per_video > 1:
-            print(f"Extracting {num_clips_per_video} clips per video...")
-        if frame_stride > 1:
-            print(f"Subsampling frames with stride {frame_stride}...")
-        if target_size:
-            print(f"Resizing clips to {target_size}...")
-
-    # Track statistics
-    failed_count = 0
-    total_clips = 0
-
-    iterator = tqdm(video_paths,
-                    desc="Loading videos") if verbose else video_paths
-
-    for video_path in iterator:
-        try:
-            # Load full video
-            video = load_video_auto(video_path,
-                                    num_frames=None,
-                                    sample_strategy='uniform')
-
-            # Sample clips from video
-            clips = sample_clips_from_video(video,
-                                            num_frames_per_clip=num_frames,
-                                            num_clips=num_clips_per_video,
-                                            strategy=clip_strategy,
-                                            frame_stride=frame_stride)
-
-            if target_size is not None:
-                resized_clips = []
-                for clip in clips:
-                    T, C, H, W = clip.shape
-                    if target_size != (H, W):
-                        # Resize to target size
-                        clip = clip.contiguous(
-                        )  # Fix non-contiguous tensors first
-                        clip_flat = clip.view(T * C, H,
-                                              W).unsqueeze(0)  # [1, T*C, H, W]
-                        clip_resized = torch.nn.functional.interpolate(
-                            clip_flat,
-                            size=target_size,
-                            mode='bilinear',
-                            align_corners=False)
-                        clip = clip_resized.squeeze(0).view(
-                            T, C, target_size[0],
-                            target_size[1])  # Back to [T, C, H, W]
-                    resized_clips.append(clip)
-                clips = resized_clips
-
-            # Yield clips one by one
-            for clip in clips:
-                yield clip
-                total_clips += 1
-
-            # Free memory
-            del video, clips
-
-        except Exception as e:
-            failed_count += 1
-            if verbose:
-                print(f"\nWarning: Failed to load {video_path}: {e}")
-            continue
-
-    # Validate
-    if total_clips == 0:
-        raise RuntimeError(f"Failed to load any videos from {directory}")
-
-    failure_rate = failed_count / len(video_paths)
-    if failure_rate > 0.1:  # More than 10% failed
-        print(
-            f"\nWARNING: {failure_rate:.1%} of videos failed to load ({failed_count}/{len(video_paths)})"
-        )
-
-    if verbose:
-        print(
-            f"\nSuccessfully loaded {total_clips} clips from {len(video_paths) - failed_count} videos"
-        )

benchmarks/scripts/run.sh

@@ -1,7 +0,0 @@
-#!/bin/bash
-
-# 1. Install missing dependency
-pip install -q opencv-python-headless transformers huggingface_hub
-
-# 2. Run FVD script
-python benchmarks/fvd/run_fvd.py

benchmarks/scripts/setup_fvd.sh

@@ -1,4 +0,0 @@
-#!/bin/bash
-
-# 1. Install missing dependency
-pip install -q opencv-python-headless

cog.yaml

@@ -0,0 +1,24 @@
+# Configuration for Cog ⚙️
+# Reference: https://cog.run/yaml
+
+build:
+  gpu: true
+  cuda: "12.1"
+  python_version: "3.10"
+  python_packages:
+    - "torch==2.4.0"
+    - "torchvision"
+    - "ninja==1.11.1.3"
+    - "transformers==4.46.1"
+    - "git+https://github.com/huggingface/diffusers.git@bf64b32652a63a1865a0528a73a13652b201698b"
+    - "accelerate==1.0.1"
+    - "safetensors==0.4.5"
+    - "peft==0.13.2"
+    - "packaging==24.2"
+    - "git+https://github.com/hao-ai-lab/FastVideo"
+
+  run:
+    - FLASH_ATTENTION_SKIP_CUDA_BUILD=TRUE pip install flash-attn --no-build-isolation
+    - curl -o /usr/local/bin/pget -L "https://github.com/replicate/pget/releases/latest/download/pget_$(uname -s)_$(uname -m)" && chmod +x /usr/local/bin/pget
+
+predict: "predict.py:Predictor"

comfyui/README.md

@@ -1,138 +0,0 @@
-# ComfyUI-FastVideo
-
-A custom node suite for ComfyUI that provides accelerated video generation using [FastVideo](https://github.com/hao-ai-labs/FastVideo). See the [blog post](https://hao-ai-lab.github.io/blogs/fastvideo/) about FastVideo V1 to learn more.
-
-## Multi-GPU Parallel Inference
-
-One of the key features ComfyUI-FastVideo brings to ComfyUI is its ability to distribute the generation workload across multiple GPUs, resulting in significantly faster inference times.
-
-![Wan2.1-I2V-14B-480P-Diffusers](./assets/wani2v.gif)
-
-Example of Wan2.1-I2V-14B-480P-Diffusers model running on 4 GPUs.
-## Features
-
-- Generate high-quality videos from text prompts and images
-- Configurable video parameters (prompt, resolution, frame count, FPS)
-- Support for multiple GPUs with tensor and sequence parallelism
-- Advanced configuration options for VAE, Text Encoder, and DIT components
-- Interruption/cancellation support for long-running generations
-
-## Installation
-
-### Requirements
-
-- [ComfyUI](https://github.com/comfyanonymous/ComfyUI)
-- CUDA-capable GPU(s) with sufficient VRAM
-
-### Install using ComfyUI Manager
-
-Coming soon!
-
-### Manual Installation
-
-#### Copy the FastVideo `comfyui` directory into your ComfyUI custom_nodes directory:
-
-```bash
-cp -r /path/to/FastVideo/comfyui /path/to/ComfyUI/custom_nodes/FastVideo
-```
-
-#### Install dependencies:
-
-Currently, the only dependency is `fastvideo`, which can be installed using pip.
-
-```bash
-pip install fastvideo
-```
-
-#### Install missing custom nodes:
-
-`ComfyUI-VideoHelperSuite`:
-
-```bash
-cd /path/to/ComfyUI/custom_nodes
-git clone https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite.git
-```
-
-If you're seeing `ImportError: libGL.so.1: cannot open shared object file: No such file or directory`,
-you may need to install ffmpeg
-
-```bash
-apt-get update && apt-get install ffmpeg
-```
-
-## Usage
-
-After installation, the following nodes will be available in the ComfyUI interface under the "fastvideo" category:
-
-- **Video Generator**: The main node for generating videos from prompts
-- **Inference Args**: Configure video generation parameters
-- **VAE Config**
-- **Text Encoder Config**
-- **DIT Config**
-- **Load Image Path**: Load images for potential conditioning
-
-You may have noticed many arguments on the nodes have 'auto' as the default value. This is because FastVideo will automatically detect the best values for these parameters based on the model and the hardware. However, you can also manually configure these parameters to get the best performance for your specific use case. We plan on releasing more optimized workflow files for different models and hardware configurations in the future.
-
-You can see what some of the default configurations are by looking at the FastVideo repo:
-- [Wan2.1-I2V-14B-480P-Diffusers](https://github.com/hao-ai-lab/FastVideo/blob/main/fastvideo/configs/wan_14B_i2v_480p_pipeline.json)
-- [FastHunyuan-diffusers](https://github.com/hao-ai-lab/FastVideo/blob/main/fastvideo/configs/fasthunyuan_t2v.json)
-
-### Node Configuration
-
-#### Video Generator
-
-- **prompt**: Text description of the video to generate
-- **output_path**: Directory where generated videos will be saved
-- **num_gpus**: Number of GPUs to use for generation
-- **model_path**: Path to the FastVideo model
-- **embedded_cfg_scale**: Classifier-free guidance scale
-- **sp_size**: Sequence parallelism size (usually should match num_gpus)
-- **tp_size**: Tensor parallelism size (usually should match num_gpus)
-- **precision**: Model precision (fp16 or bf16)
-
-`model_path takes either a model id from huggingface or a local path to a model. Models by default will be downloaded to ~/.cache/huggingface/hub/ and cached for subsequent runs.`
-
-#### Inference Args
-
-- **height/width**: Resolution of the output video
-- **num_frames**: Number of frames to generate
-- **num_inference_steps**: Number of diffusion steps per frame
-- **guidance_scale**: Classifier-free guidance scale
-- **flow_shift**: Frame flow shift parameter
-- **seed**: Random seed for reproducible generation
-- **fps**: Frames per second of the output video
-- **image_path**: Optional path to input image for conditioning (for i2v models)
-
-## Memory Management
-
-Models will remain loaded in GPU memory between runs when you only change inference arguments (such as prompt, resolution, frame count, FPS, guidance scale, etc.) or the prompt text. This allows for faster subsequent generations since the model doesn't need to be reloaded.
-
-However, if you need to change the following parameters, you will need to restart the ComfyUI server:
-- **Number of GPUs** (`num_gpus`)
-- **Model path** (`model_path`)
-- **Tensor parallelism size** (`tp_size`)
-- **Sequence parallelism size** (`sp_size`)
-
-These parameters affect the model's distribution across GPUs and require a complete reinitialization of the model pipeline.
-
-## Example workflows
-
-### Text to Video
-
-FastVideo-FastHunyuan-diffusers
-
-![FastVideo-FastHunyuan-diffusers](./assets/fasthunyuan.png)
-
-- [FastHunyuan-diffusers.json](./examples/FastHunyuan-diffusers.json)
-
-### Image to Video
-
-Wan2.1-I2V-14B-480P-Diffusers
-
-![Wan2.1-I2V-14B-480P-Diffusers](./assets/wani2v.png)
-
-- [Wan2.1-I2V-14B-480P-Diffusers.json](./examples/Wan2.1-I2V-14B-480P-Diffusers.json)
-
-## License
-
-This project is licensed under Apache 2.0.

comfyui/__init__.py

@@ -1,5 +0,0 @@
-from .video_generator.nodes import (NODE_CLASS_MAPPINGS,
-                                    NODE_DISPLAY_NAME_MAPPINGS)
-
-WEB_DIRECTORY = "./web"
-__all__ = ['NODE_CLASS_MAPPINGS', 'NODE_DISPLAY_NAME_MAPPINGS', 'WEB_DIRECTORY']

comfyui/assets/icon_simple.svg

@@ -1,6 +0,0 @@
-<svg width="160" height="93" viewBox="0 0 160 93" fill="none" xmlns="http://www.w3.org/2000/svg">
-<path d="M28.8511 91.66L57.6319 1.86368H64.5394L35.7585 91.66H28.8511Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-<path d="M15.0376 91.66L43.8185 1.86368H46.1209L17.3401 91.66H15.0376Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-<path d="M1.22217 91.66L30.003 1.86366H31.1543L2.3734 91.66H1.22217Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.15122"/>
-<path d="M71.4465 1.86483L42.666 91.6599H69.144L78.3538 58.2746H123.251L129.007 39.855H84.1099L89.866 22.5868H152.032L157.788 1.86483H71.4465Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-</svg>

comfyui/examples/FastHunyuan-diffusers.json

@@ -1,645 +0,0 @@
-{
-  "id": "23a1f065-bbba-4a8f-b144-944e1318fcbf",
-  "revision": 0,
-  "last_node_id": 8,
-  "last_link_id": 7,
-  "nodes": [
-    {
-      "id": 4,
-      "type": "VAEConfig",
-      "pos": [
-        374.2159423828125,
-        554.85888671875
-      ],
-      "size": [
-        334.080078125,
-        322
-      ],
-      "flags": {},
-      "order": 0,
-      "mode": 0,
-      "inputs": [],
-      "outputs": [
-        {
-          "name": "vae_config",
-          "type": "VAE_CONFIG",
-          "links": [
-            2
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "VAEConfig"
-      },
-      "widgets_values": [
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999
-      ],
-      "auto_widget_states": {
-        "load_encoder": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        },
-        "load_decoder": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        },
-        "tile_sample_min_height": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 256
-        },
-        "tile_sample_min_width": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 256
-        },
-        "tile_sample_min_num_frames": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 16
-        },
-        "tile_sample_stride_height": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 192
-        },
-        "tile_sample_stride_width": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 192
-        },
-        "tile_sample_stride_num_frames": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 12
-        },
-        "blend_num_frames": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 0
-        },
-        "use_tiling": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        },
-        "use_temporal_tiling": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        },
-        "use_parallel_tiling": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        }
-      }
-    },
-    {
-      "id": 5,
-      "type": "TextEncoderConfig",
-      "pos": [
-        416.4937744140625,
-        953.6171875
-      ],
-      "size": [
-        270,
-        106
-      ],
-      "flags": {},
-      "order": 1,
-      "mode": 0,
-      "inputs": [],
-      "outputs": [
-        {
-          "name": "text_encoder_config",
-          "type": "TEXT_ENCODER_CONFIG",
-          "links": [
-            7
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "TextEncoderConfig"
-      },
-      "widgets_values": [
-        -99999,
-        -99999,
-        -99999
-      ],
-      "auto_widget_states": {
-        "prefix": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": ""
-        },
-        "quant_config": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": ""
-        },
-        "lora_config": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": ""
-        }
-      }
-    },
-    {
-      "id": 6,
-      "type": "DITConfig",
-      "pos": [
-        415.1928405761719,
-        1154.1573486328125
-      ],
-      "size": [
-        270,
-        82
-      ],
-      "flags": {},
-      "order": 2,
-      "mode": 0,
-      "inputs": [],
-      "outputs": [
-        {
-          "name": "dit_config",
-          "type": "DIT_CONFIG",
-          "links": [
-            6
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "DITConfig"
-      },
-      "widgets_values": [
-        -99999,
-        -99999
-      ],
-      "auto_widget_states": {
-        "prefix": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": ""
-        },
-        "quant_config": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": ""
-        }
-      }
-    },
-    {
-      "id": 1,
-      "type": "VideoGenerator",
-      "pos": [
-        818.804931640625,
-        348.9299621582031
-      ],
-      "size": [
-        400,
-        436
-      ],
-      "flags": {},
-      "order": 4,
-      "mode": 0,
-      "inputs": [
-        {
-          "name": "inference_args",
-          "shape": 7,
-          "type": "INFERENCE_ARGS",
-          "link": 3
-        },
-        {
-          "name": "vae_config",
-          "shape": 7,
-          "type": "VAE_CONFIG",
-          "link": 2
-        },
-        {
-          "name": "text_encoder_config",
-          "shape": 7,
-          "type": "TEXT_ENCODER_CONFIG",
-          "link": 7
-        },
-        {
-          "name": "dit_config",
-          "shape": 7,
-          "type": "DIT_CONFIG",
-          "link": 6
-        }
-      ],
-      "outputs": [
-        {
-          "name": "video_path",
-          "type": "STRING",
-          "links": [
-            4
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "VideoGenerator"
-      },
-      "widgets_values": [
-        "A curious raccoon peers through a vibrant field of yellow sunflowers, its eyes wide with interest. The playful yet serene atmosphere is complemented by soft natural light filtering through the petals. Mid-shot, warm and cheerful tones.",
-        "/workspace/ComfyUI/outputs_video/",
-        2,
-        "FastVideo/FastHunyuan-diffusers",
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999,
-        -99999
-      ],
-      "auto_widget_states": {
-        "embedded_cfg_scale": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 6
-        },
-        "sp_size": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 2
-        },
-        "tp_size": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 2
-        },
-        "vae_precision": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": "fp16"
-        },
-        "vae_tiling": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        },
-        "vae_sp": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        },
-        "text_encoder_precision": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": "fp16"
-        },
-        "precision": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": "fp16"
-        },
-        "dit_cpu_offload": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        }
-      }
-    },
-    {
-      "id": 3,
-      "type": "VHS_LoadVideoPath",
-      "pos": [
-        1350.136962890625,
-        331.20361328125
-      ],
-      "size": [
-        231.8896484375,
-        286
-      ],
-      "flags": {},
-      "order": 5,
-      "mode": 0,
-      "inputs": [
-        {
-          "name": "meta_batch",
-          "shape": 7,
-          "type": "VHS_BatchManager",
-          "link": null
-        },
-        {
-          "name": "vae",
-          "shape": 7,
-          "type": "VAE",
-          "link": null
-        },
-        {
-          "name": "video",
-          "type": "STRING",
-          "widget": {
-            "name": "video"
-          },
-          "link": 4
-        }
-      ],
-      "outputs": [
-        {
-          "name": "IMAGE",
-          "type": "IMAGE",
-          "links": [
-            5
-          ]
-        },
-        {
-          "name": "frame_count",
-          "type": "INT",
-          "links": null
-        },
-        {
-          "name": "audio",
-          "type": "AUDIO",
-          "links": null
-        },
-        {
-          "name": "video_info",
-          "type": "VHS_VIDEOINFO",
-          "links": null
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "VHS_LoadVideoPath"
-      },
-      "widgets_values": {
-        "video": "",
-        "force_rate": 0,
-        "custom_width": 0,
-        "custom_height": 0,
-        "frame_load_cap": 0,
-        "skip_first_frames": 0,
-        "select_every_nth": 1,
-        "format": "Wan",
-        "videopreview": {
-          "hidden": false,
-          "paused": false,
-          "params": {
-            "filename": "",
-            "type": "path",
-            "format": "video/",
-            "force_rate": 0,
-            "custom_width": 0,
-            "custom_height": 0,
-            "frame_load_cap": 0,
-            "skip_first_frames": 0,
-            "select_every_nth": 1
-          }
-        }
-      }
-    },
-    {
-      "id": 2,
-      "type": "InferenceArgs",
-      "pos": [
-        411.46307373046875,
-        178.18182373046875
-      ],
-      "size": [
-        278.73828125,
-        298
-      ],
-      "flags": {},
-      "order": 3,
-      "mode": 0,
-      "inputs": [],
-      "outputs": [
-        {
-          "name": "inference_args",
-          "type": "INFERENCE_ARGS",
-          "links": [
-            3
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "InferenceArgs"
-      },
-      "widgets_values": [
-        720,
-        1280,
-        45,
-        6,
-        -99999,
-        -99999,
-        1025,
-        "fixed",
-        24,
-        -99999,
-        -99999
-      ],
-      "auto_widget_states": {
-        "height": {
-          "isAuto": false,
-          "value": 720,
-          "cachedValue": 720
-        },
-        "width": {
-          "isAuto": false,
-          "value": 1280,
-          "cachedValue": 1280
-        },
-        "num_frames": {
-          "isAuto": false,
-          "value": 45,
-          "cachedValue": 45
-        },
-        "num_inference_steps": {
-          "isAuto": false,
-          "value": 6,
-          "cachedValue": 6
-        },
-        "guidance_scale": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 1
-        },
-        "flow_shift": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": 17
-        },
-        "seed": {
-          "isAuto": false,
-          "value": 1025,
-          "cachedValue": 1024
-        },
-        "fps": {
-          "isAuto": false,
-          "value": 24,
-          "cachedValue": 24
-        },
-        "image_path": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": "X://insert/path/here.mp4"
-        },
-        "enable_teacache": {
-          "isAuto": true,
-          "value": -99999,
-          "cachedValue": true
-        }
-      }
-    },
-    {
-      "id": 8,
-      "type": "VHS_VideoCombine",
-      "pos": [
-        1668.3499755859375,
-        328.22625732421875
-      ],
-      "size": [
-        507.507080078125,
-        622.2227172851562
-      ],
-      "flags": {},
-      "order": 6,
-      "mode": 0,
-      "inputs": [
-        {
-          "name": "images",
-          "type": "IMAGE",
-          "link": 5
-        },
-        {
-          "name": "audio",
-          "shape": 7,
-          "type": "AUDIO",
-          "link": null
-        },
-        {
-          "name": "meta_batch",
-          "shape": 7,
-          "type": "VHS_BatchManager",
-          "link": null
-        },
-        {
-          "name": "vae",
-          "shape": 7,
-          "type": "VAE",
-          "link": null
-        }
-      ],
-      "outputs": [
-        {
-          "name": "Filenames",
-          "type": "VHS_FILENAMES",
-          "links": null
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "VHS_VideoCombine"
-      },
-      "widgets_values": {
-        "frame_rate": 24,
-        "loop_count": 0,
-        "filename_prefix": "",
-        "format": "video/h264-mp4",
-        "pix_fmt": "yuv420p",
-        "crf": 19,
-        "save_metadata": true,
-        "trim_to_audio": false,
-        "pingpong": false,
-        "save_output": false,
-        "videopreview": {
-          "hidden": false,
-          "paused": false,
-          "params": {
-            "filename": "._00003.mp4",
-            "subfolder": "",
-            "type": "temp",
-            "format": "video/h264-mp4",
-            "frame_rate": 24,
-            "workflow": "._00003.png",
-            "fullpath": "/workspace/ComfyUI/temp/._00003.mp4"
-          }
-        }
-      }
-    }
-  ],
-  "links": [
-    [
-      2,
-      4,
-      0,
-      1,
-      1,
-      "VAE_CONFIG"
-    ],
-    [
-      3,
-      2,
-      0,
-      1,
-      0,
-      "INFERENCE_ARGS"
-    ],
-    [
-      4,
-      1,
-      0,
-      3,
-      2,
-      "STRING"
-    ],
-    [
-      5,
-      3,
-      0,
-      8,
-      0,
-      "IMAGE"
-    ],
-    [
-      6,
-      6,
-      0,
-      1,
-      3,
-      "DIT_CONFIG"
-    ],
-    [
-      7,
-      5,
-      0,
-      1,
-      2,
-      "TEXT_ENCODER_CONFIG"
-    ]
-  ],
-  "groups": [],
-  "config": {},
-  "extra": {
-    "ds": {
-      "scale": 0.9090909090909091,
-      "offset": [
-        112.86678372727341,
-        -71.45635903989245
-      ]
-    },
-    "frontendVersion": "1.20.4",
-    "VHS_latentpreview": false,
-    "VHS_latentpreviewrate": 0,
-    "VHS_MetadataImage": true,
-    "VHS_KeepIntermediate": true
-  },
-  "version": 0.4
-}

comfyui/examples/Wan2.1-I2V-14B-480P-Diffusers.json

@@ -1,697 +0,0 @@
-{
-  "id": "23a1f065-bbba-4a8f-b144-944e1318fcbf",
-  "revision": 0,
-  "last_node_id": 8,
-  "last_link_id": 7,
-  "nodes": [
-    {
-      "id": 7,
-      "type": "LoadImagePath",
-      "pos": [
-        33.15385437011719,
-        191.2037353515625
-      ],
-      "size": [
-        270,
-        334
-      ],
-      "flags": {},
-      "order": 0,
-      "mode": 0,
-      "inputs": [],
-      "outputs": [
-        {
-          "name": "image_path",
-          "type": "STRING",
-          "links": [
-            1
-          ]
-        },
-        {
-          "name": "IMAGE",
-          "type": "IMAGE",
-          "links": null
-        },
-        {
-          "name": "MASK",
-          "type": "MASK",
-          "links": null
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "LoadImagePath"
-      },
-      "widgets_values": [
-        "woman.jpg",
-        "image"
-      ]
-    },
-    {
-      "id": 3,
-      "type": "VHS_LoadVideoPath",
-      "pos": [
-        1350.136962890625,
-        331.20361328125
-      ],
-      "size": [
-        231.8896484375,
-        286
-      ],
-      "flags": {},
-      "order": 6,
-      "mode": 0,
-      "inputs": [
-        {
-          "name": "meta_batch",
-          "shape": 7,
-          "type": "VHS_BatchManager",
-          "link": null
-        },
-        {
-          "name": "vae",
-          "shape": 7,
-          "type": "VAE",
-          "link": null
-        },
-        {
-          "name": "video",
-          "type": "STRING",
-          "widget": {
-            "name": "video"
-          },
-          "link": 4
-        }
-      ],
-      "outputs": [
-        {
-          "name": "IMAGE",
-          "type": "IMAGE",
-          "links": [
-            5
-          ]
-        },
-        {
-          "name": "frame_count",
-          "type": "INT",
-          "links": null
-        },
-        {
-          "name": "audio",
-          "type": "AUDIO",
-          "links": null
-        },
-        {
-          "name": "video_info",
-          "type": "VHS_VIDEOINFO",
-          "links": null
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "VHS_LoadVideoPath"
-      },
-      "widgets_values": {
-        "video": "",
-        "force_rate": 0,
-        "custom_width": 0,
-        "custom_height": 0,
-        "frame_load_cap": 0,
-        "skip_first_frames": 0,
-        "select_every_nth": 1,
-        "format": "Wan",
-        "videopreview": {
-          "hidden": false,
-          "paused": false,
-          "params": {
-            "filename": "",
-            "type": "path",
-            "format": "video/",
-            "force_rate": 0,
-            "custom_width": 0,
-            "custom_height": 0,
-            "frame_load_cap": 0,
-            "skip_first_frames": 0,
-            "select_every_nth": 1
-          }
-        }
-      }
-    },
-    {
-      "id": 8,
-      "type": "VHS_VideoCombine",
-      "pos": [
-        1668.3499755859375,
-        328.22625732421875
-      ],
-      "size": [
-        214.7587890625,
-        334
-      ],
-      "flags": {},
-      "order": 7,
-      "mode": 0,
-      "inputs": [
-        {
-          "name": "images",
-          "type": "IMAGE",
-          "link": 5
-        },
-        {
-          "name": "audio",
-          "shape": 7,
-          "type": "AUDIO",
-          "link": null
-        },
-        {
-          "name": "meta_batch",
-          "shape": 7,
-          "type": "VHS_BatchManager",
-          "link": null
-        },
-        {
-          "name": "vae",
-          "shape": 7,
-          "type": "VAE",
-          "link": null
-        }
-      ],
-      "outputs": [
-        {
-          "name": "Filenames",
-          "type": "VHS_FILENAMES",
-          "links": null
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "VHS_VideoCombine"
-      },
-      "widgets_values": {
-        "frame_rate": 24,
-        "loop_count": 0,
-        "filename_prefix": "",
-        "format": "video/h264-mp4",
-        "pix_fmt": "yuv420p",
-        "crf": 19,
-        "save_metadata": true,
-        "trim_to_audio": false,
-        "pingpong": false,
-        "save_output": false,
-        "videopreview": {
-          "hidden": false,
-          "paused": false,
-          "params": {}
-        }
-      }
-    },
-    {
-      "id": 4,
-      "type": "VAEConfig",
-      "pos": [
-        374.2159423828125,
-        554.85888671875
-      ],
-      "size": [
-        334.080078125,
-        322
-      ],
-      "flags": {},
-      "order": 1,
-      "mode": 0,
-      "inputs": [],
-      "outputs": [
-        {
-          "name": "vae_config",
-          "type": "VAE_CONFIG",
-          "links": [
-            2
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "VAEConfig"
-      },
-      "widgets_values": [
-        true,
-        true,
-        256,
-        256,
-        16,
-        192,
-        192,
-        12,
-        0,
-        true,
-        true,
-        true
-      ],
-      "auto_widget_states": {
-        "load_encoder": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        },
-        "load_decoder": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        },
-        "tile_sample_min_height": {
-          "isAuto": true,
-          "value": 256,
-          "cachedValue": 256
-        },
-        "tile_sample_min_width": {
-          "isAuto": true,
-          "value": 256,
-          "cachedValue": 256
-        },
-        "tile_sample_min_num_frames": {
-          "isAuto": true,
-          "value": 16,
-          "cachedValue": 16
-        },
-        "tile_sample_stride_height": {
-          "isAuto": true,
-          "value": 192,
-          "cachedValue": 192
-        },
-        "tile_sample_stride_width": {
-          "isAuto": true,
-          "value": 192,
-          "cachedValue": 192
-        },
-        "tile_sample_stride_num_frames": {
-          "isAuto": true,
-          "value": 12,
-          "cachedValue": 12
-        },
-        "blend_num_frames": {
-          "isAuto": true,
-          "value": 0,
-          "cachedValue": 0
-        },
-        "use_tiling": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        },
-        "use_temporal_tiling": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        },
-        "use_parallel_tiling": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        }
-      }
-    },
-    {
-      "id": 2,
-      "type": "InferenceArgs",
-      "pos": [
-        411.46307373046875,
-        178.18182373046875
-      ],
-      "size": [
-        278.73828125,
-        298
-      ],
-      "flags": {},
-      "order": 4,
-      "mode": 0,
-      "inputs": [
-        {
-          "name": "image_path",
-          "shape": 7,
-          "type": "STRING",
-          "widget": {
-            "name": "image_path"
-          },
-          "link": 1
-        }
-      ],
-      "outputs": [
-        {
-          "name": "inference_args",
-          "type": "INFERENCE_ARGS",
-          "links": [
-            3
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "InferenceArgs"
-      },
-      "widgets_values": [
-        832,
-        480,
-        45,
-        20,
-        1,
-        17,
-        1024,
-        "fixed",
-        24,
-        "X://insert/path/here.mp4",
-        true
-      ],
-      "auto_widget_states": {
-        "height": {
-          "isAuto": false,
-          "value": 832,
-          "cachedValue": 720
-        },
-        "width": {
-          "isAuto": false,
-          "value": 480,
-          "cachedValue": 1280
-        },
-        "num_frames": {
-          "isAuto": false,
-          "value": 45,
-          "cachedValue": 45
-        },
-        "num_inference_steps": {
-          "isAuto": false,
-          "value": 20,
-          "cachedValue": 6
-        },
-        "guidance_scale": {
-          "isAuto": true,
-          "value": 1,
-          "cachedValue": 1
-        },
-        "flow_shift": {
-          "isAuto": true,
-          "value": 17,
-          "cachedValue": 17
-        },
-        "seed": {
-          "isAuto": false,
-          "value": 1024,
-          "cachedValue": 1024
-        },
-        "fps": {
-          "isAuto": false,
-          "value": 24,
-          "cachedValue": 24
-        },
-        "image_path": {
-          "isAuto": true,
-          "value": "X://insert/path/here.mp4",
-          "cachedValue": "X://insert/path/here.mp4"
-        },
-        "enable_teacache": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        }
-      }
-    },
-    {
-      "id": 1,
-      "type": "VideoGenerator",
-      "pos": [
-        818.804931640625,
-        348.9299621582031
-      ],
-      "size": [
-        400,
-        436
-      ],
-      "flags": {},
-      "order": 5,
-      "mode": 0,
-      "inputs": [
-        {
-          "name": "inference_args",
-          "shape": 7,
-          "type": "INFERENCE_ARGS",
-          "link": 3
-        },
-        {
-          "name": "vae_config",
-          "shape": 7,
-          "type": "VAE_CONFIG",
-          "link": 2
-        },
-        {
-          "name": "text_encoder_config",
-          "shape": 7,
-          "type": "TEXT_ENCODER_CONFIG",
-          "link": 7
-        },
-        {
-          "name": "dit_config",
-          "shape": 7,
-          "type": "DIT_CONFIG",
-          "link": 6
-        }
-      ],
-      "outputs": [
-        {
-          "name": "video_path",
-          "type": "STRING",
-          "links": [
-            4
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "VideoGenerator"
-      },
-      "widgets_values": [
-        "A woman crying from laughter.",
-        "/workspace/ComfyUI/outputs_video/",
-        4,
-        "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers",
-        6,
-        2,
-        2,
-        "fp16",
-        true,
-        true,
-        "fp16",
-        "fp16",
-        true
-      ],
-      "auto_widget_states": {
-        "embedded_cfg_scale": {
-          "isAuto": true,
-          "value": 6,
-          "cachedValue": 6
-        },
-        "sp_size": {
-          "isAuto": true,
-          "value": 2,
-          "cachedValue": 2
-        },
-        "tp_size": {
-          "isAuto": true,
-          "value": 2,
-          "cachedValue": 2
-        },
-        "vae_precision": {
-          "isAuto": true,
-          "value": "fp16",
-          "cachedValue": "fp16"
-        },
-        "vae_tiling": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        },
-        "vae_sp": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        },
-        "text_encoder_precision": {
-          "isAuto": true,
-          "value": "fp16",
-          "cachedValue": "fp16"
-        },
-        "precision": {
-          "isAuto": true,
-          "value": "fp16",
-          "cachedValue": "fp16"
-        },
-        "dit_cpu_offload": {
-          "isAuto": true,
-          "value": true,
-          "cachedValue": true
-        }
-      }
-    },
-    {
-      "id": 5,
-      "type": "TextEncoderConfig",
-      "pos": [
-        416.4937744140625,
-        953.6171875
-      ],
-      "size": [
-        270,
-        106
-      ],
-      "flags": {},
-      "order": 2,
-      "mode": 0,
-      "inputs": [],
-      "outputs": [
-        {
-          "name": "text_encoder_config",
-          "type": "TEXT_ENCODER_CONFIG",
-          "links": [
-            7
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "TextEncoderConfig"
-      },
-      "widgets_values": [
-        "",
-        "",
-        ""
-      ],
-      "auto_widget_states": {
-        "prefix": {
-          "isAuto": true,
-          "value": "",
-          "cachedValue": ""
-        },
-        "quant_config": {
-          "isAuto": true,
-          "value": "",
-          "cachedValue": ""
-        },
-        "lora_config": {
-          "isAuto": true,
-          "value": "",
-          "cachedValue": ""
-        }
-      }
-    },
-    {
-      "id": 6,
-      "type": "DITConfig",
-      "pos": [
-        415.1928405761719,
-        1154.1573486328125
-      ],
-      "size": [
-        270,
-        82
-      ],
-      "flags": {},
-      "order": 3,
-      "mode": 0,
-      "inputs": [],
-      "outputs": [
-        {
-          "name": "dit_config",
-          "type": "DIT_CONFIG",
-          "links": [
-            6
-          ]
-        }
-      ],
-      "properties": {
-        "Node name for S&R": "DITConfig"
-      },
-      "widgets_values": [
-        "",
-        ""
-      ],
-      "auto_widget_states": {
-        "prefix": {
-          "isAuto": true,
-          "value": "",
-          "cachedValue": ""
-        },
-        "quant_config": {
-          "isAuto": true,
-          "value": "",
-          "cachedValue": ""
-        }
-      }
-    }
-  ],
-  "links": [
-    [
-      1,
-      7,
-      0,
-      2,
-      0,
-      "STRING"
-    ],
-    [
-      2,
-      4,
-      0,
-      1,
-      1,
-      "VAE_CONFIG"
-    ],
-    [
-      3,
-      2,
-      0,
-      1,
-      0,
-      "INFERENCE_ARGS"
-    ],
-    [
-      4,
-      1,
-      0,
-      3,
-      2,
-      "STRING"
-    ],
-    [
-      5,
-      3,
-      0,
-      8,
-      0,
-      "IMAGE"
-    ],
-    [
-      6,
-      6,
-      0,
-      1,
-      3,
-      "DIT_CONFIG"
-    ],
-    [
-      7,
-      5,
-      0,
-      1,
-      2,
-      "TEXT_ENCODER_CONFIG"
-    ]
-  ],
-  "groups": [],
-  "config": {},
-  "extra": {
-    "ds": {
-      "scale": 0.8264462809917354,
-      "offset": [
-        646.7950212991898,
-        66.17259910028655
-      ]
-    },
-    "frontendVersion": "1.20.4",
-    "VHS_latentpreview": false,
-    "VHS_latentpreviewrate": 0,
-    "VHS_MetadataImage": true,
-    "VHS_KeepIntermediate": true
-  },
-  "version": 0.4
-}

comfyui/video_generator/dit_config.py

@@ -1,31 +0,0 @@
-class DITConfig:
-
-    @classmethod
-    def INPUT_TYPES(cls):
-        return {
-            "optional": {
-                "prefix": ("STRING", {
-                    "default": ""
-                }),
-                "quant_config": ("STRING", {
-                    "default": ""
-                }),
-            }
-        }
-
-    @classmethod
-    def VALIDATE_INPUTS(cls, **kwargs):
-        return True
-
-    RETURN_TYPES = ("DIT_CONFIG", )
-    RETURN_NAMES = ("dit_config", )
-    FUNCTION = "set_args"
-    CATEGORY = "fastvideo"
-
-    def set_args(self, prefix, quant_config):
-        raw_args = {"prefix": prefix, "quant_config": quant_config}
-
-        # Filter out keys where value is -99999
-        args = {k: v for k, v in raw_args.items() if str(int(v)) != str(-99999)}
-
-        return (args, )

comfyui/video_generator/inference_args.py

@@ -1,89 +0,0 @@
-class InferenceArgs:
-
-    @classmethod
-    def INPUT_TYPES(cls):
-        return {
-            "optional": {
-                "height": ("INT", {
-                    "default": 720
-                }),
-                "width": ("INT", {
-                    "default": 1280
-                }),
-                "num_frames": ("INT", {
-                    "default": 45
-                }),
-                "num_inference_steps": ("INT", {
-                    "default": 6
-                }),
-                "guidance_scale": ("FLOAT", {
-                    "default": 1.0
-                }),
-                "flow_shift": ("INT", {
-                    "default": 17
-                }),
-                "seed": ("INT", {
-                    "default": 1024
-                }),
-                "fps": ("INT", {
-                    "default": 24
-                }),
-                "image_path": ("STRING", {
-                    "default": "X://insert/path/here.mp4"
-                }),
-                "enable_teacache": ([True, False], {
-                    "default": False
-                }),
-            }
-        }
-
-    @classmethod
-    def VALIDATE_INPUTS(cls, **kwargs):
-        return True
-
-    RETURN_TYPES = ("INFERENCE_ARGS", )
-    RETURN_NAMES = ("inference_args", )
-    FUNCTION = "set_args"
-    CATEGORY = "fastvideo"
-
-    def set_args(
-        self,
-        height,
-        width,
-        num_frames,
-        num_inference_steps,
-        guidance_scale,
-        flow_shift,
-        seed,
-        fps,
-        image_path,
-        enable_teacache,
-    ):
-        raw_args = {
-            "height": height,
-            "width": width,
-            "num_frames": num_frames,
-            "num_inference_steps": num_inference_steps,
-            "guidance_scale": guidance_scale,
-            "flow_shift": flow_shift,
-            "seed": seed,
-            "fps": fps,
-            "image_path": image_path,
-            "enable_teacache": enable_teacache,
-        }
-
-        # Filter out keys where value is -99999, handling different types properly
-        args = {}
-        for k, v in raw_args.items():
-            try:
-                if isinstance(v, str):
-                    if v != "-99999":
-                        args[k] = v
-                elif v != -99999:
-                    # If it's not a string, compare directly
-                    args[k] = v
-            except (ValueError, TypeError):
-                # Include any value that causes an error in comparison
-                args[k] = v
-
-        return (args, )

comfyui/video_generator/load_image.py

@@ -1,103 +0,0 @@
-import hashlib
-import os
-
-import folder_paths
-import numpy as np
-import torch
-from PIL import Image, ImageOps, ImageSequence
-
-from .node_helpers import pillow
-
-
-class LoadImagePath:
-
-    @classmethod
-    def INPUT_TYPES(s):
-        input_dir = folder_paths.get_input_directory()
-        files = [
-            f for f in os.listdir(input_dir)
-            if os.path.isfile(os.path.join(input_dir, f))
-        ]
-        files = folder_paths.filter_files_content_types(files, ["image"])
-        return {
-            "required": {
-                "image": (sorted(files), {
-                    "image_upload": True
-                })
-            },
-        }
-
-    CATEGORY = "fastvideo"
-
-    RETURN_TYPES = ("STRING", "IMAGE", "MASK")
-    RETURN_NAMES = ("image_path", "IMAGE", "MASK")
-    FUNCTION = "load_image"
-
-    def load_image(self, image):
-        image_path = folder_paths.get_annotated_filepath(image)
-
-        img = pillow(Image.open, image_path)
-
-        output_images: list[torch.Tensor] = []
-        output_masks: list[torch.Tensor] = []
-        w, h = None, None
-
-        excluded_formats = ['MPO']
-
-        for i in ImageSequence.Iterator(img):
-            processed_image = pillow(ImageOps.exif_transpose, i)
-            if processed_image is None:
-                continue
-
-            if processed_image.mode == 'I':
-                processed_image = processed_image.point(lambda i: i * (1 / 255))
-            image = processed_image.convert("RGB")
-
-            if len(output_images) == 0:
-                w = image.size[0]
-                h = image.size[1]
-
-            if image.size[0] != w or image.size[1] != h:
-                continue
-
-            image = np.array(image).astype(np.float32) / 255.0
-            image = torch.from_numpy(image)[
-                None,
-            ]
-            if 'A' in processed_image.getbands():
-                mask = np.array(processed_image.getchannel('A')).astype(
-                    np.float32) / 255.0
-                mask = 1. - torch.from_numpy(mask)
-            elif processed_image.mode == 'P' and 'transparency' in processed_image.info:
-                mask = np.array(
-                    processed_image.convert('RGBA').getchannel('A')).astype(
-                        np.float32) / 255.0
-                mask = 1. - torch.from_numpy(mask)
-            else:
-                mask = torch.zeros((64, 64), dtype=torch.float32, device="cpu")
-            output_images.append(image)
-            output_masks.append(mask.unsqueeze(0))
-
-        if len(output_images) > 1 and img.format not in excluded_formats:
-            output_image = torch.cat(output_images, dim=0)
-            output_mask = torch.cat(output_masks, dim=0)
-        else:
-            output_image = output_images[0]
-            output_mask = output_masks[0]
-
-        return (image_path, output_image, output_mask)
-
-    @classmethod
-    def IS_CHANGED(s, image):
-        image_path = folder_paths.get_annotated_filepath(image)
-        m = hashlib.sha256()
-        with open(image_path, 'rb') as f:
-            m.update(f.read())
-        return m.digest().hex()
-
-    @classmethod
-    def VALIDATE_INPUTS(s, image):
-        if not folder_paths.exists_annotated_filepath(image):
-            return "Invalid image file: {}".format(image)
-
-        return True

comfyui/video_generator/node_helpers.py

@@ -1,68 +0,0 @@
-import hashlib
-from collections.abc import Callable
-from typing import Any, TypeVar
-
-import torch
-from comfy.cli_args import args
-from PIL import ImageFile, UnidentifiedImageError
-
-T = TypeVar('T')
-
-
-def conditioning_set_values(conditioning: list[Any],
-                            values: dict[str, Any] | None = None) -> list[Any]:
-    if values is None:
-        values = {}
-    c = []
-    for t in conditioning:
-        n = [t[0], t[1].copy()]
-        for k in values:
-            n[1][k] = values[k]
-        c.append(n)
-
-    return c
-
-
-def pillow(fn: Callable[[Any], T], arg: Any) -> T:
-    prev_value = None
-    try:
-        x = fn(arg)
-    except (OSError, UnidentifiedImageError, ValueError
-            ):  #PIL issues #4472 and #2445, also fixes ComfyUI issue #3416
-        prev_value = ImageFile.LOAD_TRUNCATED_IMAGES
-        ImageFile.LOAD_TRUNCATED_IMAGES = True
-        x = fn(arg)
-    finally:
-        if prev_value is not None:
-            ImageFile.LOAD_TRUNCATED_IMAGES = prev_value
-    return x
-
-
-def hasher() -> Callable[[], Any]:
-    hashfuncs = {
-        "md5": hashlib.md5,
-        "sha1": hashlib.sha1,
-        "sha256": hashlib.sha256,
-        "sha512": hashlib.sha512
-    }
-    return hashfuncs[args.default_hashing_function]
-
-
-def string_to_torch_dtype(string: str) -> torch.dtype | None:
-    if string == "fp32":
-        return torch.float32
-    if string == "fp16":
-        return torch.float16
-    if string == "bf16":
-        return torch.bfloat16
-    return None
-
-
-def image_alpha_fix(destination: torch.Tensor,
-                    source: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor]:
-    if destination.shape[-1] < source.shape[-1]:
-        source = source[..., :destination.shape[-1]]
-    elif destination.shape[-1] > source.shape[-1]:
-        destination = torch.nn.functional.pad(destination, (0, 1))
-        destination[..., -1] = 1.0
-    return destination, source

comfyui/video_generator/nodes.py

@@ -1,24 +0,0 @@
-from .dit_config import DITConfig
-from .inference_args import InferenceArgs
-from .load_image import LoadImagePath
-from .text_encoder_config import TextEncoderConfig
-from .vae_config import VAEConfig
-from .video_generator import VideoGenerator
-
-NODE_CLASS_MAPPINGS = {
-    "VideoGenerator": VideoGenerator,
-    "InferenceArgs": InferenceArgs,
-    "VAEConfig": VAEConfig,
-    "TextEncoderConfig": TextEncoderConfig,
-    "DITConfig": DITConfig,
-    "LoadImagePath": LoadImagePath
-}
-
-NODE_DISPLAY_NAME_MAPPINGS = {
-    "VideoGenerator": "Video Generator",
-    "InferenceArgs": "Inference Args",
-    "VAEConfig": "VAE Config",
-    "TextEncoderConfig": "Text Encoder Config",
-    "DITConfig": "DIT Config",
-    "LoadImagePath": "Load Image Path"
-}

comfyui/video_generator/text_encoder_config.py

@@ -1,38 +0,0 @@
-class TextEncoderConfig:
-
-    @classmethod
-    def INPUT_TYPES(cls):
-        return {
-            "optional": {
-                "prefix": ("STRING", {
-                    "default": ""
-                }),
-                "quant_config": ("STRING", {
-                    "default": ""
-                }),
-                "lora_config": ("STRING", {
-                    "default": ""
-                }),
-            }
-        }
-
-    @classmethod
-    def VALIDATE_INPUTS(cls, **kwargs):
-        return True
-
-    RETURN_TYPES = ("TEXT_ENCODER_CONFIG", )
-    RETURN_NAMES = ("text_encoder_config", )
-    FUNCTION = "set_args"
-    CATEGORY = "fastvideo"
-
-    def set_args(self, prefix, quant_config, lora_config):
-        raw_args = {
-            "prefix": prefix,
-            "quant_config": quant_config,
-            "lora_config": lora_config
-        }
-
-        # Filter out keys where value is -99999
-        args = {k: v for k, v in raw_args.items() if str(int(v)) != str(-99999)}
-
-        return (args, )

comfyui/video_generator/vae_config.py

@@ -1,88 +0,0 @@
-class VAEConfig:
-
-    @classmethod
-    def INPUT_TYPES(cls):
-        return {
-            "optional": {
-                "load_encoder": ([True, False], {
-                    "default": True
-                }),
-                "load_decoder": ([True, False], {
-                    "default": True
-                }),
-                "tile_sample_min_height": ("INT", {
-                    "default": 256
-                }),
-                "tile_sample_min_width": ("INT", {
-                    "default": 256
-                }),
-                "tile_sample_min_num_frames": ("INT", {
-                    "default": 16
-                }),
-                "tile_sample_stride_height": ("INT", {
-                    "default": 192
-                }),
-                "tile_sample_stride_width": ("INT", {
-                    "default": 192
-                }),
-                "tile_sample_stride_num_frames": ("INT", {
-                    "default": 12
-                }),
-                "blend_num_frames": ("INT", {
-                    "default": 0
-                }),
-                "use_tiling": ([True, False], {
-                    "default": True
-                }),
-                "use_temporal_tiling": ([True, False], {
-                    "default": True
-                }),
-                "use_parallel_tiling": ([True, False], {
-                    "default": True
-                }),
-            }
-        }
-
-    @classmethod
-    def VALIDATE_INPUTS(cls, **kwargs):
-        return True
-
-    RETURN_TYPES = ("VAE_CONFIG", )
-    RETURN_NAMES = ("vae_config", )
-    FUNCTION = "set_args"
-    CATEGORY = "fastvideo"
-
-    def set_args(
-        self,
-        load_encoder,
-        load_decoder,
-        tile_sample_min_height,
-        tile_sample_min_width,
-        tile_sample_min_num_frames,
-        tile_sample_stride_height,
-        tile_sample_stride_width,
-        tile_sample_stride_num_frames,
-        blend_num_frames,
-        use_tiling,
-        use_temporal_tiling,
-        use_parallel_tiling,
-    ):
-        raw_args = {
-            "load_encoder": load_encoder,
-            "load_decoder": load_decoder,
-            "tile_sample_min_height": tile_sample_min_height,
-            "tile_sample_min_width": tile_sample_min_width,
-            "tile_sample_min_num_frames": tile_sample_min_num_frames,
-            "tile_sample_stride_height": tile_sample_stride_height,
-            "tile_sample_stride_width": tile_sample_stride_width,
-            "tile_sample_stride_num_frames": tile_sample_stride_num_frames,
-            "blend_num_frames": blend_num_frames,
-            "use_tiling": use_tiling,
-            "use_temporal_tiling": use_temporal_tiling,
-            "use_parallel_tiling": use_parallel_tiling,
-        }
-
-        # Filter out any value explicitly set to -99999
-        args = {k: v for k, v in raw_args.items() if str(int(v)) != str(-99999)}
-
-        return (args, )

comfyui/video_generator/video_generator.py

@@ -1,315 +0,0 @@
-from __future__ import annotations
-
-import glob
-import os
-import signal
-import sys
-import threading
-import time
-from typing import Any
-
-from comfy.model_management import processing_interrupted
-
-from fastvideo import PipelineConfig
-from fastvideo import VideoGenerator as FastVideoGenerator
-
-sys.path.insert(
-    0,
-    os.path.dirname(
-        os.path.dirname(
-            os.path.dirname(os.path.dirname(os.path.abspath(__file__))))))
-
-
-# Custom exception for interruption
-class GenerationInterruptedException(Exception):
-    pass
-
-
-# Custom exception for interruption that ComfyUI will recognize
-class GenerationCancelledException(Exception):
-
-    def __init__(self,
-                 message: str = "Generation was cancelled by user") -> None:
-        self.message = message
-        super().__init__(self.message)
-
-
-def update_config_from_args(config: Any, args_dict: dict[str, Any]) -> None:
-    """
-    Update configuration object from arguments dictionary.
-    
-    Args:
-        config: The configuration object to update
-        args_dict: Dictionary containing arguments
-    """
-    for key, value in args_dict.items():
-        if hasattr(config, key) and value is not None:
-            if key == "text_encoder_precisions" and isinstance(value, list):
-                setattr(config, key, tuple(value))
-            else:
-                setattr(config, key, value)
-
-
-class VideoGenerator:
-
-    @classmethod
-    def INPUT_TYPES(s):
-        return {
-            "required": {
-                "prompt": ("STRING", {
-                    "multiline":
-                    True,
-                    "default":
-                    "A ripe orange tumbles gently from a tree and lands on the head of a lounging capybara, "
-                    "who blinks slowly in response. The moment is quietly humorous and oddly serene, framed by "
-                    "lush green foliage and dappled sunlight. Mid-shot, warm and whimsical tones."
-                }),
-                "output_path": ("STRING", {
-                    "default": "/workspace/ComfyUI/outputs_video/"
-                }),
-                "num_gpus": ("INT", {
-                    "default": 2,
-                    "min": 1,
-                    "max": 16
-                }),
-                "model_path": ("STRING", {
-                    "default": "FastVideo/FastHunyuan-diffusers"
-                })
-            },
-            "optional": {
-                "inference_args": ("INFERENCE_ARGS", ),
-                "embedded_cfg_scale": ("FLOAT", {
-                    "default": 6.0
-                }),
-                "sp_size": ("INT", {
-                    "default": 2
-                }),
-                "tp_size": ("INT", {
-                    "default": 2
-                }),
-                "vae_config": ("VAE_CONFIG", ),
-                "vae_precision": (["fp16", "bf16"], {
-                    "default": "fp16"
-                }),
-                "vae_tiling": ([True, False], {
-                    "default": True
-                }),
-                "vae_sp": ([True, False], {
-                    "default": False
-                }),
-                "text_encoder_config": ("TEXT_ENCODER_CONFIG", ),
-                "text_encoder_precision": (["fp16", "bf16"], {
-                    "default": "fp16"
-                }),
-                "dit_config": ("DIT_CONFIG", ),
-                "precision": (["fp16", "bf16"], {
-                    "default": "fp16"
-                }),
-                "dit_cpu_offload": ([True, False], {
-                    "default": False
-                }),
-            }
-        }
-
-    @classmethod
-    def VALIDATE_INPUTS(cls, **kwargs):
-        return True
-
-    RETURN_TYPES = ("STRING", )
-    RETURN_NAMES = ("video_path", )
-    FUNCTION = "launch_inference"
-    CATEGORY = "fastvideo"
-
-    generator: FastVideoGenerator | None = None
-    _interrupt_thread: threading.Thread | None = None
-    _generation_active: bool = False
-    _generation_interrupted: bool = False
-    _interrupt_event: threading.Event = threading.Event()
-    _generation_thread: threading.Thread | None = None
-    _generation_result: str | None = None
-    _generation_exception: Exception | None = None
-
-    def _monitor_for_interruption(self):
-        """Background thread that monitors for interruption requests"""
-        time.sleep(2)  # Give the generation thread time to send execute_forward
-
-        while self._generation_active and not self._interrupt_event.is_set():
-            if processing_interrupted():
-                print("Video generation interrupted by user")
-                self._generation_interrupted = True
-
-                # Try to send interrupt signal to worker processes
-                if self.generator is not None and hasattr(
-                        self.generator, 'executor'):
-                    try:
-                        # The MultiprocExecutor has a workers attribute
-                        if hasattr(self.generator.executor, 'workers'):
-                            for worker in self.generator.executor.workers:
-                                if worker.is_alive():
-                                    os.kill(worker.pid, signal.SIGINT)
-                            print("Interrupt signal sent to worker processes")
-                    except Exception as e:
-                        print(f"Error sending interrupt signal: {e}")
-
-                # Set the interrupt event to notify other threads
-                self._interrupt_event.set()
-                break
-            time.sleep(0.5)
-
-    def _run_generation(self, prompt: str, output_path: str,
-                        inference_args: dict[str, Any]) -> None:
-        """Thread function to run the generation"""
-        try:
-            if self.generator is not None:
-                self.generator.generate_video(prompt=prompt,
-                                              output_path=output_path,
-                                              **inference_args)
-                self._generation_result = os.path.join(output_path,
-                                                       f"{prompt[:100]}.mp4")
-            else:
-                raise RuntimeError("Generator is not initialized")
-        except Exception as e:
-            self._generation_exception = e
-            self._interrupt_event.set()
-
-    def load_output_video(self, output_dir):
-        video_extensions = ["*.mp4", "*.avi", "*.mov", "*.mkv"]
-        video_files = []
-
-        for ext in video_extensions:
-            video_files.extend(glob.glob(os.path.join(output_dir, ext)))
-
-        if not video_files:
-            print("No video files found in output directory: %s", output_dir)
-            return ""
-
-        video_files.sort()
-        return video_files[0]
-
-    def launch_inference(
-        self,
-        prompt,
-        output_path,
-        num_gpus,
-        model_path,
-        embedded_cfg_scale,
-        sp_size,
-        tp_size,
-        vae_precision,
-        vae_tiling,
-        vae_sp,
-        text_encoder_precision,
-        precision,
-        inference_args=None,
-        vae_config=None,
-        text_encoder_config=None,
-        dit_config=None,
-        dit_cpu_offload=None,
-    ):
-        print('Running FastVideo inference')
-
-        # Reset interruption flag and event
-        self._generation_interrupted = False
-        self._interrupt_event.clear()
-        self._generation_result = None
-        self._generation_exception = None
-
-        # Load pipeline config from model path
-        pipeline_config = PipelineConfig.from_pretrained(model_path)
-        print('pipeline_config', pipeline_config)
-
-        # Update configs with provided config dictionaries
-        if dit_config is not None:
-            update_config_from_args(pipeline_config.dit_config, dit_config)
-
-        if vae_config is not None:
-            update_config_from_args(pipeline_config.vae_config, vae_config)
-
-        if text_encoder_config is not None:
-            update_config_from_args(pipeline_config.text_encoder_configs,
-                                    text_encoder_config)
-
-        # Update top-level pipeline config with remaining arguments
-        raw_pipeline_args = {}
-        if embedded_cfg_scale is not None:
-            raw_pipeline_args['embedded_cfg_scale'] = embedded_cfg_scale
-        if precision is not None:
-            raw_pipeline_args['precision'] = precision
-        if vae_precision is not None:
-            raw_pipeline_args['vae_precision'] = vae_precision
-        if vae_tiling is not None:
-            raw_pipeline_args['vae_tiling'] = vae_tiling
-        if vae_sp is not None:
-            raw_pipeline_args['vae_sp'] = vae_sp
-        if text_encoder_precision is not None:
-            raw_pipeline_args['text_encoder_precision'] = text_encoder_precision
-
-        # Filter out any value explicitly set to -99999 (auto values)
-        pipeline_args = {
-            k: v
-            for k, v in raw_pipeline_args.items() if str(int(v)) != str(-99999)
-        }
-
-        update_config_from_args(pipeline_config, pipeline_args)
-
-        raw_generation_args = {}
-        if num_gpus is not None:
-            raw_generation_args['num_gpus'] = num_gpus
-        if tp_size is not None:
-            raw_generation_args['tp_size'] = tp_size
-        if sp_size is not None:
-            raw_generation_args['sp_size'] = sp_size
-        if dit_cpu_offload is not None:
-            raw_generation_args['dit_cpu_offload'] = dit_cpu_offload
-
-        generation_args = {
-            k: v
-            for k, v in raw_generation_args.items()
-            if str(int(v)) != str(-99999)
-        }
-
-        if self.generator is None:
-            print('generation_args', generation_args)
-            print('pipeline_config', pipeline_config)
-            self.generator = FastVideoGenerator.from_pretrained(
-                model_path=model_path,
-                **generation_args,
-                pipeline_config=pipeline_config)
-
-        print('inference_args', inference_args)
-
-        # Start a thread to run the generation
-        self._generation_thread = threading.Thread(target=self._run_generation,
-                                                   args=(prompt, output_path,
-                                                         inference_args),
-                                                   daemon=True)
-        self._generation_thread.start()
-
-        # Start a background thread to monitor for interruptions
-        self._generation_active = True
-        self._interrupt_thread = threading.Thread(
-            target=self._monitor_for_interruption, daemon=True)
-        self._interrupt_thread.start()
-
-        # Wait for either completion or interruption
-        while self._generation_thread.is_alive(
-        ) and not self._interrupt_event.is_set():
-            self._generation_thread.join(timeout=0.5)
-
-        self._generation_active = False
-        if self._interrupt_thread:
-            self._interrupt_thread.join(timeout=1.0)
-            self._interrupt_thread = None
-
-        if self._generation_interrupted:
-            print("Video generation was cancelled by user")
-            raise GenerationCancelledException()
-        elif self._generation_exception:
-            # Re-raise the exception from the generation thread
-            raise self._generation_exception
-        elif self._generation_result:
-            return (self._generation_result, )
-        else:
-            # This shouldn't happen, but just in case
-            print("Generation completed but no result was produced")
-            raise Exception("Generation failed to produce a result")

comfyui/web/js/FastVideo.core.js

@@ -1,593 +0,0 @@
-import { app } from '../../../scripts/app.js'
-
-function chainCallback(object, property, callback) {
-    if (object == undefined) {
-        console.error("Tried to add callback to non-existent object");
-        return;
-    }
-    if (property in object && object[property]) {
-        const callback_orig = object[property];
-        object[property] = function () {
-            const r = callback_orig.apply(this, arguments);
-            return callback.apply(this, arguments) ?? r;
-        };
-    } else {
-        object[property] = callback;
-    }
-}
-
-function drawAutoAnnotated(ctx, node, widget_width, y, H) {
-    const litegraph_base = LiteGraph;
-    const show_text = app.canvas.ds.scale >= 0.5;
-    const margin = 15;
-
-    const autoTextWidth = 30;
-    const autoTextRightMargin = 5;
-
-    ctx.textAlign = 'left';
-    ctx.strokeStyle = litegraph_base.WIDGET_OUTLINE_COLOR;
-    ctx.fillStyle = litegraph_base.WIDGET_BGCOLOR;
-
-    ctx.beginPath();
-    if (show_text && ctx.roundRect) {
-        ctx.roundRect(margin, y, widget_width - margin * 2, H, [H * 0.5]);
-    } else {
-        ctx.rect(margin, y, widget_width - margin * 2, H);
-    }
-    ctx.fill();
-
-    if (show_text) {
-        if (!this.disabled) ctx.stroke();
-        const isAuto = this.isAuto === true;
-
-        ctx.save();
-        if (isAuto) {
-            ctx.fillStyle = litegraph_base.WIDGET_TEXT_COLOR;
-            ctx.strokeStyle = litegraph_base.WIDGET_TEXT_COLOR;
-        } else {
-            ctx.fillStyle = litegraph_base.WIDGET_SECONDARY_TEXT_COLOR;
-            ctx.strokeStyle = litegraph_base.WIDGET_SECONDARY_TEXT_COLOR;
-        }
-
-        // Position for the cog
-        const cogX = widget_width - autoTextRightMargin - autoTextWidth - 6;
-        const cogY = y + H * 0.5;
-        const cogRadius = 6;
-        const toothLength = 2;
-        const numTeeth = 8;
-        const holeRadius = 2; // Radius of the center hole
-
-        // Draw the cog
-        ctx.beginPath();
-        ctx.arc(cogX, cogY, cogRadius - toothLength, 0, Math.PI * 2);
-        ctx.fill();
-
-        // Draw the center hole (by clearing it)
-        ctx.beginPath();
-        ctx.arc(cogX, cogY, holeRadius, 0, Math.PI * 2);
-        ctx.fillStyle = litegraph_base.WIDGET_BGCOLOR;
-        ctx.fill();
-
-        // Reset fill style for the teeth
-        if (isAuto) {
-            ctx.fillStyle = litegraph_base.WIDGET_TEXT_COLOR;
-        } else {
-            ctx.fillStyle = litegraph_base.WIDGET_SECONDARY_TEXT_COLOR;
-        }
-
-        // Draw teeth
-        ctx.beginPath();
-        for (let i = 0; i < numTeeth; i++) {
-            const angle = (i / numTeeth) * Math.PI * 2;
-            const innerX = cogX + (cogRadius - toothLength) * Math.cos(angle);
-            const innerY = cogY + (cogRadius - toothLength) * Math.sin(angle);
-            const outerX = cogX + cogRadius * Math.cos(angle);
-            const outerY = cogY + cogRadius * Math.sin(angle);
-
-            ctx.moveTo(innerX, innerY);
-            ctx.lineTo(outerX, outerY);
-        }
-        ctx.lineWidth = 2;
-        ctx.stroke();
-
-        ctx.restore();
-
-        // Draw label
-        ctx.fillStyle = litegraph_base.WIDGET_SECONDARY_TEXT_COLOR;
-        const label = this.label || this.name;
-        if (label != null) {
-            ctx.fillText(label, margin * 2 + 5, y + H * 0.7);
-        }
-
-        // Draw value
-        ctx.textAlign = 'right';
-        const text = isAuto ? "auto" : this.displayValue();
-        ctx.fillStyle = isAuto ? litegraph_base.WIDGET_SECONDARY_TEXT_COLOR : litegraph_base.WIDGET_TEXT_COLOR;
-        ctx.fillText(text, widget_width - autoTextRightMargin - autoTextWidth - 15, y + H * 0.7);
-
-        // Draw increment/decrement buttons if not in AUTO mode and not a string widget
-        if (!isAuto && !this.disabled && this.config[0] !== "FVAUTOSTRING") {
-            // Draw decrement button (left triangle)
-            ctx.fillStyle = litegraph_base.WIDGET_TEXT_COLOR;
-            ctx.beginPath();
-            ctx.moveTo(margin + 16, y + 5);
-            ctx.lineTo(margin + 6, y + H * 0.5);
-            ctx.lineTo(margin + 16, y + H - 5);
-            ctx.fill();
-
-            // Draw increment button (right triangle)
-            ctx.beginPath();
-            ctx.moveTo(widget_width - margin - 16, y + 5);
-            ctx.lineTo(widget_width - margin - 6, y + H * 0.5);
-            ctx.lineTo(widget_width - margin - 16, y + H - 5);
-            ctx.fill();
-        }
-    }
-}
-
-function mouseAutoAnnotated(event, [x, y], node) {
-    const widget_width = node.size[0];
-    const margin = 15;
-    const H = 20; // Widget height
-
-    const autoTextWidth = 30;
-    const autoTextRightMargin = 5;
-
-    const cogRadius = 6;
-
-    if (this.isAuto) {
-        if (event.type === "pointerup" || event.type === "mouseup") {
-            const cogX = widget_width - autoTextRightMargin - autoTextWidth - 6;
-            const cogLeftEdge = cogX - cogRadius;
-            const cogRightEdge = cogX + cogRadius;
-
-            if (x > cogLeftEdge && x < cogRightEdge) {
-                this.isAuto = false;
-                this.value = this.cachedValue !== undefined ? this.cachedValue : (this.options.default || 0);
-
-                if (this.callback) {
-                    this.callback(this.value);
-                }
-                node.graph.setDirtyCanvas(true, false);
-            }
-        }
-
-        // Block ALL events in auto mode except cog clicks
-        event.preventDefault?.();
-        event.stopPropagation?.();
-        event.stopImmediatePropagation?.();
-        return true; // Always return true to indicate event was handled
-    }
-
-    // Determine if clicking on increment/decrement buttons
-    const delta = this.config[0] === "FVAUTOSTRING" ? 0 :
-        (x < 40 ? -1 : x > widget_width - 48 ? 1 : 0);
-
-    if (event.type === "pointerdown" || event.type === "mousedown") {
-        // ComfyUI appears to intercept pointerdown events, so this code path is never reached
-        console.log("pointerdown received (unexpected)");
-        return false;
-    } else if (event.type === "pointerup" || event.type === "mouseup") {
-        // Stop event propagation to prevent double handling
-        event.preventDefault?.();
-        event.stopPropagation?.();
-        event.stopImmediatePropagation?.();
-
-        const cogX = widget_width - autoTextRightMargin - autoTextWidth - 6;
-        const cogLeftEdge = widget_width - autoTextRightMargin - autoTextWidth - 6 - cogRadius;
-        const cogRightEdge = widget_width - autoTextRightMargin - autoTextWidth - 6 + cogRadius;
-
-        if (x > cogLeftEdge && x < cogRightEdge) {
-            this.isAuto = !this.isAuto;
-
-            if (this.isAuto) {
-                this.cachedValue = this.value;
-                this.value = -99999;
-            } else {
-                this.value = this.cachedValue !== undefined ? this.cachedValue : (this.options.default || 0);
-            }
-
-            if (this.callback) {
-                this.callback(this.value);
-            }
-
-            node.graph.setDirtyCanvas(true, false);
-            return true;
-        }
-
-        // If in auto mode and NOT clicking the cog, block all other interactions
-        if (this.isAuto) {
-            return true;
-        }
-
-        // Handle increment/decrement buttons if not in auto mode
-        if (delta !== 0 && !this.isAuto) {
-            if (this.config[0] === "FVAUTOCOMBO") {
-                const options = this.options.values || [];
-                if (options.length === 0) return true;
-
-                let currentIndex = -1;
-                for (let i = 0; i < options.length; i++) {
-                    const optValue = typeof options[i] === 'object' ? options[i].value : options[i];
-                    if (optValue == this.value || String(optValue) === String(this.value)) {
-                        currentIndex = i;
-                        break;
-                    }
-                }
-
-                if (currentIndex === -1) {
-                    currentIndex = 0;
-                }
-
-                let newIndex = currentIndex + delta;
-                if (newIndex < 0) {
-                    newIndex = options.length - 1;
-                } else if (newIndex >= options.length) {
-                    newIndex = 0;
-                }
-
-                const newOption = options[newIndex];
-                this.value = typeof newOption === 'object' ? newOption.value : newOption;
-
-                if (this.callback) {
-                    this.callback(this.value);
-                }
-
-                node.graph.setDirtyCanvas(true, false);
-                return true;
-            } else {
-                let v = parseFloat(this.value);
-                const increment = delta * 0.1 * (this.options.step || 1);
-
-                v += increment;
-
-                // Apply min/max constraints
-                if (this.options.min != null) {
-                    v = Math.max(this.options.min, v);
-                }
-                if (this.options.max != null) {
-                    v = Math.min(this.options.max, v);
-                }
-
-                // Round to precision or to integer
-                if (this.config[0] === "FVAUTOINT") {
-                    v = Math.round(v);
-                } else if (this.options.precision !== undefined) {
-                    const precision = Math.pow(10, this.options.precision);
-                    v = Math.round(v * precision) / precision;
-                }
-
-                this.value = v;
-
-                if (this.callback) {
-                    this.callback(this.value);
-                }
-
-                node.graph.setDirtyCanvas(true, false);
-                return true;
-            }
-        }
-
-        if (delta === 0 && !this.isAuto) {
-            if (this.config[0] === "FVAUTOCOMBO") {
-                const options = this.options.values || [];
-
-                // Create menu items
-                const menuItems = options.map(opt => {
-                    const value = typeof opt === 'object' ? opt.value : opt;
-                    const label = typeof opt === 'object' ? opt.label : opt.toString();
-
-                    return {
-                        content: label,
-                        callback: () => {
-                            this.value = value;
-
-                            if (this.callback) {
-                                this.callback(this.value);
-                            }
-
-                            node.graph.setDirtyCanvas(true, false);
-                        }
-                    };
-                });
-
-                new LiteGraph.ContextMenu(menuItems, {
-                    event: event,
-                    title: null,
-                    callback: null,
-                    extra: node
-                });
-
-                return true;
-            } else if (this.config[0] === "FVAUTOSTRING") {
-                const d_callback = (v) => {
-                    this.value = v;
-
-                    if (this.callback) {
-                        this.callback(this.value);
-                    }
-
-                    node.graph.setDirtyCanvas(true, false);
-                };
-
-                const dialog = app.canvas.prompt(
-                    'Value',
-                    this.value,
-                    d_callback,
-                    event
-                );
-
-                return true;
-            } else {
-                // For numeric widgets, show input dialog
-                const d_callback = (v) => {
-                    this.value = this.parseValue?.(v) ?? Number(v);
-
-                    // Apply min/max constraints
-                    if (this.options.min != null) {
-                        this.value = Math.max(this.options.min, this.value);
-                    }
-                    if (this.options.max != null) {
-                        this.value = Math.min(this.options.max, this.value);
-                    }
-
-                    // Round to precision or to integer
-                    if (this.config[0] === "FVAUTOINT") {
-                        this.value = Math.round(this.value);
-                    } else if (this.options.precision !== undefined) {
-                        const precision = Math.pow(10, this.options.precision);
-                        this.value = Math.round(this.value * precision) / precision;
-                    }
-
-                    if (this.callback) {
-                        this.callback(this.value);
-                    }
-
-                    node.graph.setDirtyCanvas(true, false);
-                };
-
-                const dialog = app.canvas.prompt(
-                    'Value',
-                    this.value,
-                    d_callback,
-                    event
-                );
-
-                return true;
-            }
-        }
-
-        return true;
-    }
-
-    return false;
-}
-
-function makeAutoAnnotated(widget, inputData) {
-    const original = {
-        callback: widget.callback,
-        type: widget.type,
-        value: widget.value
-    };
-
-    // Add AUTO properties to the widget
-    Object.assign(widget, {
-        type: "BOOLEAN",
-        draw: drawAutoAnnotated,
-        mouse: mouseAutoAnnotated,
-        onMouse: null, // Explicitly disable original onMouse handler
-
-        isAuto: true,
-        cachedValue: widget.value,
-        config: inputData,
-        options: Object.assign({}, inputData[1], widget.options),
-        original: original,  // Store original properties for reference
-
-        // Disable other potential mouse handlers with no-op functions
-        onClick: function () {
-            return false;
-        },
-        onPointerUp: function () {
-            return false;
-        },
-        onPointerDown: function () {
-            return false;
-        },
-        onMouseUp: function () {
-            return false;
-        },
-        onMouseDown: function () {
-            return false;
-        },
-
-        computeSize(width) {
-            return [width, 20];
-        },
-        displayValue: function () {
-            if (this.config[0] === "FVAUTOINT") {
-                return Math.round(this.value).toString();
-            }
-            if (this.config[0] === "FVAUTOCOMBO") {
-                return this.value;
-            }
-            if (this.config[0] === "FVAUTOSTRING") {
-                return this.value;
-            }
-            // For FLOAT values, check if it's actually an integer
-            if (Number.isInteger(this.value)) {
-                return this.value.toString();
-            }
-            return this.value.toFixed(this.options.precision || 2);
-        },
-        parseValue: function (v) {
-            if (this.config[0] === "FVAUTOSTRING") {
-                return v;
-            }
-            if (typeof v === "string") {
-                return parseFloat(v);
-            }
-            return v;
-        },
-        serializeValue: function () {
-            // Return special value for AUTO mode
-            return this.isAuto ? -99999 : this.value;
-        },
-        deserializeValue: function (data) {
-            if (data === -99999) {
-                this.isAuto = true;
-                this.value = -99999;
-            } else {
-                this.isAuto = false;
-                this.value = data;
-                this.cachedValue = data;
-            }
-        }
-    });
-
-    // Override callback to handle AUTO mode
-    widget.callback = function (v) {
-        if (this.isAuto) {
-            return; // Don't call the original callback in AUTO mode
-        }
-        const result = original.callback?.call(this, v);
-        return result;
-    };
-
-    // Override any potential click handlers
-    const originalOnClick = widget.onClick;
-    if (originalOnClick) {
-        widget.onClick = function (...args) {
-            if (this.isAuto) {
-                return false;
-            }
-            return originalOnClick.call(this, ...args);
-        };
-    }
-
-    return widget;
-}
-
-app.registerExtension({
-    name: "FastVideo.AutoWidgets",
-
-    async beforeRegisterNodeDef(nodeType, nodeData, app) {
-        if (nodeData?.name == "VideoGenerator" || nodeData?.name === "InferenceArgs" || nodeData?.name === "VAEConfig" ||
-            nodeData?.name === "TextEncoderConfig" || nodeData?.name === "DITConfig") {
-            // Add serialization support
-            chainCallback(nodeType.prototype, "onSerialize", function (info) {
-                if (!this.widgets) {
-                    return;
-                }
-                // Ensure widgets_values exists
-                if (!info.widgets_values) {
-                    info.widgets_values = {};
-                }
-
-                // Store AUTO widget states in a separate property
-                if (!info.auto_widget_states) {
-                    info.auto_widget_states = {};
-                }
-
-                // Handle AUTO widgets specially
-                for (const w of this.widgets) {
-                    if (w.type === "BOOLEAN" && w.isAuto !== undefined) {
-                        // Store the serialized value (for Python node)
-                        info.widgets_values[w.name] = w.serializeValue();
-
-                        // Store the full state (for UI restoration)
-                        info.auto_widget_states[w.name] = {
-                            isAuto: w.isAuto,
-                            value: w.value,
-                            cachedValue: w.cachedValue
-                        };
-                    }
-                }
-            });
-
-            // Add deserialization support
-            chainCallback(nodeType.prototype, "onConfigure", function (info) {
-                if (!this.widgets) {
-                    return;
-                }
-
-                // First, restore from widgets_values (for backward compatibility)
-                if (info.widgets_values && Array.isArray(info.widgets_values)) {
-                    for (let i = 0; i < this.widgets.length && i < info.widgets_values.length; i++) {
-                        const w = this.widgets[i];
-                        const value = info.widgets_values[i];
-
-                        if (w.type === "BOOLEAN" && w.isAuto !== undefined) {
-                            w.deserializeValue(value);
-                        }
-                    }
-                }
-
-                // Then, restore full state if available
-                if (info.auto_widget_states) {
-                    for (const w of this.widgets) {
-                        if (w.type === "BOOLEAN" && w.isAuto !== undefined && w.name in info.auto_widget_states) {
-                            const state = info.auto_widget_states[w.name];
-
-                            w.isAuto = state.isAuto;
-                            w.cachedValue = state.cachedValue;
-                            w.value = state.isAuto ? -99999 : state.value;
-
-                            w.callback?.(w.value);
-                        }
-                    }
-                }
-
-                // Force a redraw
-                this.graph?.setDirtyCanvas(true, true);
-            });
-
-
-            // Override addInput to handle AUTO widgets
-            chainCallback(nodeType.prototype, "onNodeCreated", function () {
-                // Convert any existing widgets to AUTO widgets if needed
-                let new_widgets = [];
-                const intWidgetNames = ["sp_size", "tp_size", "height", "width", "num_frames", "num_inference_steps", "flow_shift", "seed", "fps", "scale_factor",
-                    "tile_sample_min_height", "tile_sample_min_width", "tile_sample_min_num_frames", "tile_sample_stride_height", "tile_sample_stride_width",
-                    "tile_sample_stride_num_frames", "blend_num_frames"
-                ]
-                const floatWidgetNames = ["embedded_cfg_scale", "guidance_scale"]
-                const comboWidgetNames = ["vae_tiling", "vae_precision", "vae_sp", "text_encoder_precision", "precision",
-                    "load_encoder", "load_decoder", "use_tiling", "use_temporal_tiling", "use_parallel_tiling", "dit_cpu_offload", "enable_teacache"
-                ]
-                const stringWidgetNames = ["prefix", "quant_config", "lora_config", "image_path"]
-
-                if (this.widgets) {
-                    for (let w of this.widgets) {
-                        if (intWidgetNames.includes(w.name)) {
-                            new_widgets.push(makeAutoAnnotated(w, ["FVAUTOINT", { "default": 0 }]));
-                        } else if (floatWidgetNames.includes(w.name)) {
-                            new_widgets.push(makeAutoAnnotated(w, ["FVAUTOFLOAT", { "default": 0 }]));
-                        } else if (comboWidgetNames.includes(w.name)) {
-                            new_widgets.push(makeAutoAnnotated(w, ["FVAUTOCOMBO", { "default": 0 }]));
-                        } else if (stringWidgetNames.includes(w.name)) {
-                            new_widgets.push(makeAutoAnnotated(w, ["FVAUTOSTRING", { "default": "" }]));
-                        } else {
-                            new_widgets.push(w);
-                        }
-                    }
-                    this.widgets = new_widgets;
-
-                    const autoWidgets = this.widgets.filter(w => w.type === "BOOLEAN" && w.isAuto !== undefined);
-                }
-
-                this.graph?.setDirtyCanvas(true, true);
-            });
-        }
-    },
-
-    async init() {
-        // Force a redraw of all nodes when the extension initializes
-        if (app.graph) {
-            setTimeout(() => {
-                app.graph.setDirtyCanvas(true, true);
-            }, 1000);
-        }
-    }
-});
-
-console.log("FastVideo.core.js loaded");

csrc/sliding_tile_attention/MANIFEST.in

@@ -0,0 +1,2 @@
+recursive-include tk *
+include config.py

csrc/sliding_tile_attention/README.md

@@ -0,0 +1,68 @@
+
+
+# Sliding Tile Atteniton Kernel
+
+
+## Installation
+We test our code on Pytorch 2.5.0 and CUDA>=12.4. Currently we only have implementation on H100.
+First, install C++20 for ThunderKittens:
+
+```bash
+sudo apt update
+sudo apt install gcc-11 g++-11
+
+sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
+
+sudo apt update
+sudo apt install clang-11
+```
+Install STA:
+```bash
+export CUDA_HOME=/usr/local/cuda-12.4
+export PATH=${CUDA_HOME}/bin:${PATH} 
+export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
+git submodule update --init --recursive
+python setup.py install
+```
+
+## Usage
+
+```python
+from st_attn import sliding_tile_attention
+# assuming video size (T, H, W) = (30, 48, 80), text tokens = 256 with padding. 
+# q, k, v: [batch_size, num_heads, seq_length, head_dim], seq_length = T*H*W + 256
+# a tile is a cube of size (6, 8, 8)
+# window_size in tiles: [(window_t, window_h, window_w), (..)...]. For example, window size (3, 3, 3) means a query can attend to (3x6, 3x8, 3x8) = (18, 24, 24) tokens out of the total 30x48x80 video.
+# text_length: int ranging from 0 to 256
+# If your attention contains text token (Hunyuan)
+out = sliding_tile_attention(q, k, v, window_size, text_length)
+# If your attention does not contain text token (StepVideo)
+out = sliding_tile_attention(q, k, v, window_size, 0, False)
+
+```
+
+
+## Test
+```bash
+python test/test_sta.py
+```
+
+## How Does STA Work?
+We give a demo for 2D STA with window size (6,6) operating on a (10, 10) image. 
+
+
+https://github.com/user-attachments/assets/f3b6dd79-7b43-4b60-a0fa-3d6495ec5747
+
+## Why is STA Fast?
+2D/3D Sliding Window Attention (SWA) creates many mixed blocks in the attention map. Even though mixed blocks have less output value,a mixed block is significantly slower than a dense block due to the GPU-unfriendly masking operation. 
+
+STA removes mixed blocks.
+
+
+<div align="center">
+<img src=../../assets/sliding_tile_attn_map.png width="80%"/>
+</div>
+
+## Acknowledgement
+
+We learned or reuse code from FlexAtteniton, NATEN, and ThunderKittens.

csrc/sliding_tile_attention/config.py

@@ -0,0 +1,15 @@
+### ADD TO THIS TO REGISTER NEW KERNELS
+sources = {
+    'attn': {
+        'source_files': {
+            'h100': 'st_attn/st_attn_h100.cu'  # define these source files for each GPU target desired.
+        }
+    }
+}
+
+### WHICH KERNELS DO WE WANT TO BUILD?
+# (oftentimes during development work you don't need to redefine them all.)
+kernels = ['attn']
+
+### WHICH GPU TARGET DO WE WANT TO BUILD FOR?
+target = 'h100'

csrc/sliding_tile_attention/setup.py

@@ -0,0 +1,76 @@
+import os
+import subprocess
+
+from config import kernels, sources, target
+from setuptools import find_packages, setup
+from torch.utils.cpp_extension import BuildExtension, CUDAExtension
+
+target = target.lower()
+
+# Package metadata
+PACKAGE_NAME = "st_attn"
+VERSION = "0.0.4"
+AUTHOR = "Hao AI Lab"
+DESCRIPTION = "Sliding Tile Atteniton Kernel Used in FastVideo"
+URL = "https://github.com/hao-ai-lab/FastVideo/tree/main/csrc/sliding_tile_attention"
+
+# Set environment variables
+tk_root = os.getenv('THUNDERKITTENS_ROOT', os.path.abspath(os.path.join(os.getcwd(), 'tk/')))
+python_include = subprocess.check_output(['python', '-c',
+                                          "import sysconfig; print(sysconfig.get_path('include'))"]).decode().strip()
+torch_include = subprocess.check_output([
+    'python', '-c',
+    "import torch; from torch.utils.cpp_extension import include_paths; print(' '.join(['-I' + p for p in include_paths()]))"
+]).decode().strip()
+print('st_attn root:', tk_root)
+print('Python include:', python_include)
+print('Torch include directories:', torch_include)
+
+# CUDA flags
+cuda_flags = [
+    '-DNDEBUG', '-Xcompiler=-Wno-psabi', '-Xcompiler=-fno-strict-aliasing', '--expt-extended-lambda',
+    '--expt-relaxed-constexpr', '-forward-unknown-to-host-compiler', '--use_fast_math', '-std=c++20', '-O3',
+    '-Xnvlink=--verbose', '-Xptxas=--verbose', '-Xptxas=--warn-on-spills', f'-I{tk_root}/include',
+    f'-I{tk_root}/prototype', f'-I{python_include}', '-DTORCH_COMPILE'
+] + torch_include.split()
+cpp_flags = ['-std=c++20', '-O3']
+
+if target == 'h100':
+    cuda_flags.append('-DKITTENS_HOPPER')
+    cuda_flags.append('-arch=sm_90a')
+else:
+    raise ValueError(f'Target {target} not supported')
+
+source_files = ['st_attn.cpp']
+for k in kernels:
+    if target not in sources[k]['source_files']:
+        raise KeyError(f'Target {target} not found in source files for kernel {k}')
+    if isinstance(sources[k]['source_files'][target], list):
+        source_files.extend(sources[k]['source_files'][target])
+    else:
+        source_files.append(sources[k]['source_files'][target])
+    cpp_flags.append(f'-DTK_COMPILE_{k.replace(" ", "_").upper()}')
+
+setup(name=PACKAGE_NAME,
+      version=VERSION,
+      author=AUTHOR,
+      description=DESCRIPTION,
+      url=URL,
+      packages=find_packages(),
+      ext_modules=[
+          CUDAExtension('st_attn_cuda',
+                        sources=source_files,
+                        extra_compile_args={
+                            'cxx': cpp_flags,
+                            'nvcc': cuda_flags
+                        },
+                        libraries=['cuda'])
+      ],
+      cmdclass={'build_ext': BuildExtension},
+      classifiers=[
+          "Programming Language :: Python :: 3",
+          "Environment :: GPU :: NVIDIA CUDA :: 12",
+          "License :: OSI Approved :: Apache Software License",
+      ],
+      python_requires='>=3.10',
+      install_requires=["torch>=2.5.0"])

csrc/sliding_tile_attention/st_attn.cpp

@@ -0,0 +1,24 @@
+#include <torch/extension.h>
+#include <ATen/ATen.h>
+
+#include <vector>
+#include <cuda_fp16.h>
+#include <cuda_bf16.h>
+
+#include <cuda_runtime.h>
+
+
+#ifdef TK_COMPILE_ATTN
+extern torch::Tensor sta_forward(
+    torch::Tensor q, torch::Tensor k, torch::Tensor v, torch::Tensor o, int kernel_t_size, int kernel_w_size, int kernel_h_size, int text_length, bool process_text, bool has_text, int kernel_aspect_ratio_flag
+); 
+#endif
+
+
+PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
+    m.doc() = "Sliding Block Attention Kernels"; // optional module docstring
+
+#ifdef TK_COMPILE_ATTN
+    m.def("sta_fwd",  torch::wrap_pybind_function(sta_forward), "sliding tile attention, assuming tile size is (6,8,8)");
+#endif
+}

csrc/sliding_tile_attention/st_attn/__init__.py

@@ -0,0 +1,46 @@
+import math
+
+import torch
+from st_attn_cuda import sta_fwd
+
+
+def sliding_tile_attention(q_all, k_all, v_all, window_size, text_length, has_text=True, img_latent_shape='30*48*80'):
+    seq_length = q_all.shape[2]
+    img_latent_shape_mapping = {
+        '30x48x80':1,
+        '36x48x48':2,
+        '18x48x80':3, 
+    }
+    if has_text:
+        assert q_all.shape[
+            2] >= 115200, "STA currently only supports video with latent size (30, 48, 80), which is 117 frames x 768 x 1280 pixels"
+        assert q_all.shape[1] == len(window_size), "Number of heads must match the number of window sizes"
+        target_size = math.ceil(seq_length / 384) * 384
+        pad_size = target_size - seq_length
+        if pad_size > 0:
+            q_all = torch.cat([q_all, q_all[:, :, -pad_size:]], dim=2)
+            k_all = torch.cat([k_all, k_all[:, :, -pad_size:]], dim=2)
+            v_all = torch.cat([v_all, v_all[:, :, -pad_size:]], dim=2)
+    else:
+        if img_latent_shape == '36x48x48': # Stepvideo 204x768x68
+            assert q_all.shape[2] == 82944
+        elif img_latent_shape == '18x48x80': # Wan 69x768x1280
+            assert q_all.shape[2] == 69120
+        else:
+            raise ValueError(f"Unsupported {img_latent_shape}, current shape is {q_all.shape}, only support '36x48x48' for Stepvideo and '18x48x80' for Wan")
+
+    kernel_aspect_ratio_flag = img_latent_shape_mapping[img_latent_shape]
+    hidden_states = torch.empty_like(q_all)
+    # This for loop is ugly. but it is actually quite efficient. The sequence dimension alone can already oversubscribe SMs
+    for head_index, (t_kernel, h_kernel, w_kernel) in enumerate(window_size):
+        for batch in range(q_all.shape[0]):
+            q_head, k_head, v_head, o_head = (q_all[batch:batch + 1, head_index:head_index + 1],
+                                              k_all[batch:batch + 1,
+                                                    head_index:head_index + 1], v_all[batch:batch + 1,
+                                                                                      head_index:head_index + 1],
+                                              hidden_states[batch:batch + 1, head_index:head_index + 1])
+
+            _ = sta_fwd(q_head, k_head, v_head, o_head, t_kernel, h_kernel, w_kernel, text_length, False, has_text, kernel_aspect_ratio_flag)
+    if has_text:
+        _ = sta_fwd(q_all, k_all, v_all, hidden_states, 3, 3, 3, text_length, True, True, kernel_aspect_ratio_flag)
+    return hidden_states[:, :, :seq_length]

csrc/sliding_tile_attention/st_attn/st_attn_h100.cu

@@ -0,0 +1,831 @@
+// # Define TORCH_COMPILE macro
+
+#include "kittens.cuh"
+#include <cooperative_groups.h>
+#include <iostream>
+#include <stdio.h>
+
+#define CLAMP(value, min, max) ((value) < (min) ? (min) : ((value) > (max) ? (max) : (value)))
+#define ABS(x) ((x) < 0 ? -(x) : (x))
+
+constexpr int CONSUMER_WARPGROUPS = (3); 
+constexpr int PRODUCER_WARPGROUPS = (1); 
+constexpr int NUM_WARPGROUPS      = (CONSUMER_WARPGROUPS+PRODUCER_WARPGROUPS); 
+constexpr int NUM_WORKERS         = (NUM_WARPGROUPS*kittens::WARPGROUP_WARPS); 
+
+using namespace kittens;
+namespace cg = cooperative_groups;
+
+template<int D> struct fwd_attend_ker_tile_dims {};
+template<> struct fwd_attend_ker_tile_dims<64> {
+    constexpr static int tile_width = (64);
+    constexpr static int qo_height  = (4*16);
+    constexpr static int kv_height  = (8*16);
+    constexpr static int stages     = (4); 
+};
+template<> struct fwd_attend_ker_tile_dims<128> {
+    constexpr static int tile_width = (128);
+    constexpr static int qo_height  = (4*16);
+    constexpr static int kv_height  = (8*16);
+    constexpr static int stages     = (2); 
+};
+
+template<int D> struct fwd_globals {
+    using q_tile    =         st_bf<fwd_attend_ker_tile_dims<D>::qo_height, fwd_attend_ker_tile_dims<D>::tile_width>;
+    using k_tile    =         st_bf<fwd_attend_ker_tile_dims<D>::kv_height, fwd_attend_ker_tile_dims<D>::tile_width>;
+    using v_tile    =         st_bf<fwd_attend_ker_tile_dims<D>::kv_height, fwd_attend_ker_tile_dims<D>::tile_width>;
+    using l_col_vec = col_vec<st_fl<fwd_attend_ker_tile_dims<D>::qo_height, fwd_attend_ker_tile_dims<D>::tile_width>>;
+    using o_tile    =         st_bf<fwd_attend_ker_tile_dims<D>::qo_height, fwd_attend_ker_tile_dims<D>::tile_width>;
+
+    using q_gl = gl<bf16,  -1, -1, -1, -1, q_tile>;
+    using k_gl = gl<bf16,  -1, -1, -1, -1, k_tile>;
+    using v_gl = gl<bf16,  -1, -1, -1, -1, v_tile>;
+    using l_gl = gl<float, -1, -1, -1, -1, l_col_vec>;
+    using o_gl = gl<bf16,  -1, -1, -1, -1, o_tile>;
+
+    q_gl q;
+    k_gl k;
+    v_gl v;
+    l_gl l;
+    o_gl o;
+
+    const int N; 
+    const int text_L;
+    const int hr;
+};
+
+
+template<int D, bool is_causal, bool text_q, bool text_kv, int DT, int DH, int DW, int CT, int CH, int CW>
+__global__  __launch_bounds__((NUM_WORKERS)*kittens::WARP_THREADS, 1)
+void fwd_attend_ker(const __grid_constant__ fwd_globals<D> g) {
+    extern __shared__ int __shm[]; 
+    tma_swizzle_allocator al((int*)&__shm[0]);
+    int warpid = kittens::warpid(), warpgroupid = warpid/kittens::WARPGROUP_WARPS;
+
+    using K = fwd_attend_ker_tile_dims<D>;
+
+    using q_tile    =         st_bf<K::qo_height, K::tile_width>;
+    using k_tile    =         st_bf<K::kv_height, K::tile_width>;
+    using v_tile    =         st_bf<K::kv_height, K::tile_width>;
+    using l_col_vec = col_vec<st_fl<K::qo_height, K::tile_width>>;
+    using o_tile    =         st_bf<K::qo_height, K::tile_width>;
+    
+    q_tile    (&q_smem)[CONSUMER_WARPGROUPS] = al.allocate<q_tile, CONSUMER_WARPGROUPS>();
+    k_tile    (&k_smem)[K::stages]           = al.allocate<k_tile, K::stages          >();
+    v_tile    (&v_smem)[K::stages]           = al.allocate<v_tile, K::stages          >();
+    l_col_vec (&l_smem)[CONSUMER_WARPGROUPS] = al.allocate<l_col_vec, CONSUMER_WARPGROUPS>();
+    auto      (*o_smem)                      = reinterpret_cast<o_tile(*)>(q_smem);
+    int img_kv_blocks;
+    int kv_blocks   = g.N / (K::kv_height);
+    if constexpr (text_kv) {
+        img_kv_blocks = kv_blocks - 3;
+    } else {
+        img_kv_blocks = kv_blocks;
+    }
+    int kv_head_idx = blockIdx.y / g.hr;
+    int seq_idx;
+    if constexpr (text_q) {
+        seq_idx = CT * CH * CW * 6.0 + blockIdx.x * CONSUMER_WARPGROUPS;
+    } else {
+        seq_idx = blockIdx.x * CONSUMER_WARPGROUPS; 
+    }
+    __shared__ kittens::semaphore qsmem_semaphore, k_smem_arrived[K::stages], v_smem_arrived[K::stages], compute_done[K::stages];
+    if (threadIdx.x == 0) { 
+        init_semaphore(qsmem_semaphore, 0, 1); 
+        for(int j = 0; j < K::stages; j++) {
+            init_semaphore(k_smem_arrived[j], 0, 1); 
+            init_semaphore(v_smem_arrived[j], 0, 1); 
+            init_semaphore(compute_done[j], CONSUMER_WARPGROUPS, 0); 
+        }
+
+        tma::expect_bytes(qsmem_semaphore, sizeof(q_smem));
+
+        for (int wg = 0; wg < CONSUMER_WARPGROUPS; wg++) {
+            coord<q_tile> q_tile_idx = {blockIdx.z, blockIdx.y, (seq_idx) + wg, 0};
+            tma::load_async(q_smem[wg], g.q, q_tile_idx, qsmem_semaphore);
+        }
+
+        if constexpr (text_q){
+            for (int j = 0; j < K::stages - 1; j++) {
+                coord<k_tile> kv_tile_idx = {blockIdx.z, kv_head_idx, j, 0};
+                tma::expect_bytes(k_smem_arrived[j], sizeof(k_tile));
+                tma::load_async(k_smem[j], g.k, kv_tile_idx, k_smem_arrived[j]);
+                tma::expect_bytes(v_smem_arrived[j], sizeof(v_tile));
+                tma::load_async(v_smem[j], g.v, kv_tile_idx, v_smem_arrived[j]);
+            }
+        } else {
+            int qt = seq_idx / 6 / (CH * CW);
+            int qh = (seq_idx / 6) % (CH * CW) / CW;
+            int qw = (seq_idx / 6) % CW;
+            qt = CLAMP(qt, DT, CT-DT-1);
+            qh = CLAMP(qh, DH, CH-DH-1);
+            qw = CLAMP(qw, DW, CW-DW-1);
+            int count = 0;
+            int j = 0;
+            while (count < K::stages - 1) {
+                int kt = j / 3 / (CH * CW);
+                int kh = (j / 3) % (CH * CW) / CW;
+                int kw = (j / 3) % CW;
+                bool mask = (ABS(qt - kt) <= DT) && (ABS(qh - kh) <= DH) && (ABS(qw - kw) <= DW);
+                if (mask){
+                    coord<k_tile> kv_tile_idx = {blockIdx.z, kv_head_idx, j, 0};
+                    tma::expect_bytes(k_smem_arrived[count], sizeof(k_tile));
+                    tma::load_async(k_smem[count], g.k, kv_tile_idx, k_smem_arrived[count]);
+                    tma::expect_bytes(v_smem_arrived[count], sizeof(v_tile));
+                    tma::load_async(v_smem[count], g.v, kv_tile_idx, v_smem_arrived[count]);
+                    count += 1;
+                }
+                j += 1;
+            }
+        }
+    }
+    __syncthreads(); 
+
+    int pipe_idx = K::stages - 1; 
+    
+    if(warpgroupid == NUM_WARPGROUPS-1) {
+        warpgroup::decrease_registers<32>();      
+        
+        int kv_iters; 
+        if constexpr (is_causal) {
+            kv_iters = (seq_idx * (K::qo_height/kittens::TILE_ROW_DIM<bf16>)) - 1 + (CONSUMER_WARPGROUPS * (K::qo_height/kittens::TILE_ROW_DIM<bf16>)); 
+            kv_iters = ((kv_iters / (K::kv_height/kittens::TILE_ROW_DIM<bf16>)) == 0) ? (0) : ((kv_iters / (K::kv_height/kittens::TILE_ROW_DIM<bf16>)) - 1);
+        }
+        else { kv_iters = kv_blocks-2;}
+
+        if(warpid == NUM_WORKERS-4) {
+            if constexpr (text_q){
+                for (auto kv_idx = pipe_idx - 1; kv_idx <= kv_iters; kv_idx++) {
+                    coord<k_tile> kv_tile_idx = {blockIdx.z, kv_head_idx, kv_idx + 1, 0};
+                    tma::expect_bytes(k_smem_arrived[(kv_idx+1)%K::stages], sizeof(k_tile));
+                    tma::load_async(k_smem[(kv_idx+1)%K::stages], g.k, kv_tile_idx, k_smem_arrived[(kv_idx+1)%K::stages]);
+                    tma::expect_bytes(v_smem_arrived[(kv_idx+1)%K::stages], sizeof(v_tile));
+                    tma::load_async(v_smem[(kv_idx+1)%K::stages], g.v, kv_tile_idx, v_smem_arrived[(kv_idx+1)%K::stages]);
+                    kittens::wait(compute_done[(kv_idx)%K::stages], (kv_idx/K::stages)%2);
+                }
+            } else {
+                int qt = seq_idx / 6 / (CH * CW);
+                int qh = (seq_idx / 6) % (CH * CW) / CW;
+                int qw = (seq_idx / 6) % CW;
+                qt = CLAMP(qt, DT, CT-DT-1);
+                qh = CLAMP(qh, DH, CH-DH-1);
+                qw = CLAMP(qw, DW, CW-DW-1);
+                int k_t_min = CLAMP(qt-DT, 0, CT-1);
+                int k_t_max = CLAMP(qt+DT, 0, CT-1);
+                int k_h_min = CLAMP(qh-DH, 0, CH-1);
+                int k_h_max = CLAMP(qh+DH, 0, CH-1);
+                int k_w_min = CLAMP(qw-DW, 0, CW-1);
+                int k_w_max = CLAMP(qw+DW, 0, CW-1);
+                int count = 0;
+                for (int kt = k_t_min; kt <= k_t_max; kt++) {
+                    for (int kh = k_h_min; kh <= k_h_max; kh++) {
+                        for (int kw = k_w_min; kw <= k_w_max; kw++) {
+                            for (int j = 0; j <= 2; j++){
+                                if (count >= K::stages - 1) {
+                                    int index = ((kt * (CH * CW)) + (kh * CW) + kw) * 3 + j;
+                                    coord<k_tile> kv_tile_idx = {blockIdx.z, kv_head_idx, index, 0};
+                                    tma::expect_bytes(k_smem_arrived[count%K::stages], sizeof(k_tile));
+                                    tma::load_async(k_smem[count%K::stages], g.k, kv_tile_idx, k_smem_arrived[count%K::stages]);
+                                    tma::expect_bytes(v_smem_arrived[count%K::stages], sizeof(v_tile));
+                                    tma::load_async(v_smem[count%K::stages], g.v, kv_tile_idx, v_smem_arrived[count%K::stages]);
+                                    kittens::wait(compute_done[(count - 1)%K::stages], ((count - 1)/K::stages)%2);
+                                    count += 1;
+                                } else {
+                                    count += 1;
+                                }
+                            }
+                        }
+                    }
+                }
+                // for text 
+                for (int index = img_kv_blocks; index < kv_blocks; index++) {
+                    coord<k_tile> kv_tile_idx = {blockIdx.z, kv_head_idx, index, 0};
+                    tma::expect_bytes(k_smem_arrived[count%K::stages], sizeof(k_tile));
+                    tma::load_async(k_smem[count%K::stages], g.k, kv_tile_idx, k_smem_arrived[count%K::stages]);
+                    tma::expect_bytes(v_smem_arrived[count%K::stages], sizeof(v_tile));
+                    tma::load_async(v_smem[count%K::stages], g.v, kv_tile_idx, v_smem_arrived[count%K::stages]);
+                    kittens::wait(compute_done[(count - 1)%K::stages], ((count - 1)/K::stages)%2);
+                    count += 1;
+                }
+            }
+
+
+        }
+    }
+    else {
+        warpgroup::increase_registers<160>();
+
+        rt_fl<16, K::kv_height>  att_block;
+        rt_bf<16, K::kv_height>  att_block_mma;
+        rt_fl<16, K::tile_width> o_reg;
+        
+        col_vec<rt_fl<16, K::kv_height>> max_vec, norm_vec, max_vec_last_scaled, max_vec_scaled;
+        
+        neg_infty(max_vec);
+        zero(norm_vec);
+        zero(o_reg);
+
+        int kv_iters; 
+        if constexpr (is_causal) {
+            kv_iters = (seq_idx * 4) - 1 + (CONSUMER_WARPGROUPS * 4);
+            kv_iters = (kv_iters/8);
+        }
+        else if constexpr (text_q){ 
+            // the last three kv blocks are for text, we process them separately
+            kv_iters = img_kv_blocks - 1;
+        } else {
+            kv_iters = CLAMP(DT*2+1, 1, CT) * CLAMP(DH*2+1, 1, CH) * CLAMP(DW*2+1, 1, CW) * 3 - 1 ; 
+        }
+
+        kittens::wait(qsmem_semaphore, 0);
+        for (auto kv_idx = 0; kv_idx <= kv_iters; kv_idx++) {
+
+            kittens::wait(k_smem_arrived[(kv_idx)%K::stages], (kv_idx/K::stages)%2);
+            warpgroup::mm_ABt(att_block, q_smem[warpgroupid], k_smem[(kv_idx)%K::stages]);
+            
+            copy(max_vec_last_scaled, max_vec);
+            if constexpr (D == 64) { mul(max_vec_last_scaled, max_vec_last_scaled, 1.44269504089f*0.125f); }
+            else                   { mul(max_vec_last_scaled, max_vec_last_scaled, 1.44269504089f*0.08838834764f); }
+            
+            warpgroup::mma_async_wait();
+
+            row_max(max_vec, att_block, max_vec);
+            
+            if constexpr (D == 64) { 
+                mul(att_block, att_block,    1.44269504089f*0.125f); 
+                mul(max_vec_scaled, max_vec, 1.44269504089f*0.125f);
+            }
+            else                   { 
+                mul(att_block, att_block,    1.44269504089f*0.08838834764f); 
+                mul(max_vec_scaled, max_vec, 1.44269504089f*0.08838834764f);
+            }
+
+            sub_row(att_block, att_block, max_vec_scaled);
+            exp2(att_block, att_block);
+            sub(max_vec_last_scaled, max_vec_last_scaled, max_vec_scaled);
+            exp2(max_vec_last_scaled,       max_vec_last_scaled);
+            mul(norm_vec,            norm_vec,     max_vec_last_scaled);
+            row_sum(norm_vec,  att_block, norm_vec);
+            add(att_block, att_block, 0.f);
+            copy(att_block_mma, att_block); 
+            mul_row(o_reg, o_reg, max_vec_last_scaled); 
+
+            kittens::wait(v_smem_arrived[(kv_idx)%K::stages], (kv_idx/K::stages)%2); 
+
+            warpgroup::mma_AB(o_reg, att_block_mma, v_smem[(kv_idx)%K::stages]);
+            warpgroup::mma_async_wait();
+
+            if(warpgroup::laneid() == 0) arrive(compute_done[(kv_idx)%K::stages], 1);
+        }
+        // the last three kv blocks are for text, we process them separately
+        if constexpr(text_kv) {
+            for (auto kv_idx = kv_iters + 1; kv_idx <= kv_iters + 3; kv_idx++) {
+
+                kittens::wait(k_smem_arrived[(kv_idx)%K::stages], (kv_idx/K::stages)%2);
+                warpgroup::mm_ABt(att_block, q_smem[warpgroupid], k_smem[(kv_idx)%K::stages]);
+                
+                copy(max_vec_last_scaled, max_vec);
+                if constexpr (D == 64) { mul(max_vec_last_scaled, max_vec_last_scaled, 1.44269504089f*0.125f); }
+                else                   { mul(max_vec_last_scaled, max_vec_last_scaled, 1.44269504089f*0.08838834764f); }
+                
+                warpgroup::mma_async_wait();
+                // apply non-pad mask
+                int offset = g.text_L - (kv_idx - (kv_iters + 1)) * K::kv_height;
+                // printf("k_idx_start: %d, k_idx_end: %d, text_end: %d, offset: %d\n", k_idx_start, k_idx_end, text_end, offset);
+                right_fill(att_block, att_block, offset, base_types::constants<float>::neg_infty());
+
+
+                row_max(max_vec, att_block, max_vec);
+                
+                if constexpr (D == 64) { 
+                    mul(att_block, att_block,    1.44269504089f*0.125f); 
+                    mul(max_vec_scaled, max_vec, 1.44269504089f*0.125f);
+                }
+                else                   { 
+                    mul(att_block, att_block,    1.44269504089f*0.08838834764f); 
+                    mul(max_vec_scaled, max_vec, 1.44269504089f*0.08838834764f);
+                }
+
+                sub_row(att_block, att_block, max_vec_scaled);
+                exp2(att_block, att_block);
+                sub(max_vec_last_scaled, max_vec_last_scaled, max_vec_scaled);
+                exp2(max_vec_last_scaled,       max_vec_last_scaled);
+                mul(norm_vec,            norm_vec,     max_vec_last_scaled);
+                row_sum(norm_vec,  att_block, norm_vec);
+                add(att_block, att_block, 0.f);
+                copy(att_block_mma, att_block); 
+                mul_row(o_reg, o_reg, max_vec_last_scaled); 
+
+                kittens::wait(v_smem_arrived[(kv_idx)%K::stages], (kv_idx/K::stages)%2); 
+
+                warpgroup::mma_AB(o_reg, att_block_mma, v_smem[(kv_idx)%K::stages]);
+                warpgroup::mma_async_wait();
+
+                if(warpgroup::laneid() == 0) arrive(compute_done[(kv_idx)%K::stages], 1);
+            }
+        }
+
+        div_row(o_reg, o_reg, norm_vec);
+        warpgroup::store(o_smem[warpgroupid], o_reg); 
+        warpgroup::sync(warpgroupid+4);
+
+        if (warpid % 4 == 0) {
+            coord<o_tile> o_tile_idx = {blockIdx.z, blockIdx.y, (seq_idx) + warpgroupid, 0};
+            tma::store_async(g.o, o_smem[warpgroupid], o_tile_idx);
+        }
+
+        mul(max_vec_scaled,   max_vec_scaled, 0.69314718056f);
+        log(norm_vec, norm_vec);
+        add(norm_vec, norm_vec, max_vec_scaled);
+
+        if constexpr (D == 64) { mul(norm_vec, norm_vec, -8.0f); }
+        else                   { mul(norm_vec, norm_vec, -11.313708499f); }
+    
+        warpgroup::store(l_smem[warpgroupid], norm_vec);
+        warpgroup::sync(warpgroupid+4);
+
+        if (warpid % 4 == 0) {
+            coord<l_col_vec> tile_idx = {blockIdx.z, blockIdx.y, 0, (seq_idx) + warpgroupid};
+            tma::store_async(g.l, l_smem[warpgroupid], tile_idx);
+        }
+        tma::store_async_wait();
+    }
+}
+
+
+
+#include "pyutils/torch_helpers.cuh"
+#include <ATen/cuda/CUDAContext.h>
+#include <iostream>
+
+torch::Tensor 
+sta_forward(torch::Tensor q, torch::Tensor k, torch::Tensor v, torch::Tensor o, int kernel_t_size, int kernel_h_size, int kernel_w_size, int text_length, bool process_text, bool has_text, int kernel_aspect_ratio_flag)
+{
+    CHECK_INPUT(q);
+    CHECK_INPUT(k);
+    CHECK_INPUT(v);
+
+    auto batch    = q.size(0);
+    auto seq_len  = q.size(2); 
+    auto head_dim = q.size(3);  
+    auto qo_heads = q.size(1);
+    auto kv_heads = k.size(1);
+
+    // check to see that these dimensions match for all inputs
+    TORCH_CHECK(q.size(0) == batch, "Q batch dimension - idx 0 - must match for all inputs");
+    TORCH_CHECK(k.size(0) == batch, "K batch dimension - idx 0 - must match for all inputs");
+    TORCH_CHECK(v.size(0) == batch, "V batch dimension - idx 0 - must match for all inputs");
+
+    TORCH_CHECK(q.size(2) == seq_len, "Q sequence length dimension - idx 2 - must match for all inputs");
+    TORCH_CHECK(k.size(2) == seq_len, "K sequence length dimension - idx 2 - must match for all inputs");
+    TORCH_CHECK(v.size(2) == seq_len, "V sequence length dimension - idx 2 - must match for all inputs");
+
+    TORCH_CHECK(q.size(3) == head_dim, "Q head dimension - idx 3 - must match for all non-vector inputs");
+    TORCH_CHECK(k.size(3) == head_dim, "K head dimension - idx 3 - must match for all non-vector inputs");
+    TORCH_CHECK(v.size(3) == head_dim, "V head dimension - idx 3 - must match for all non-vector inputs");
+
+    TORCH_CHECK(qo_heads >= kv_heads, "QO heads must be greater than or equal to KV heads");
+    TORCH_CHECK(qo_heads % kv_heads == 0, "QO heads must be divisible by KV heads");
+    TORCH_CHECK(q.size(1) == qo_heads, "QO head dimension - idx 1 - must match for all inputs");
+    TORCH_CHECK(k.size(1) == kv_heads, "KV head dimension - idx 1 - must match for all inputs");
+    TORCH_CHECK(v.size(1) == kv_heads, "KV head dimension - idx 1 - must match for all inputs");  
+
+    auto hr = qo_heads / kv_heads;
+
+    c10::BFloat16* q_ptr = q.data_ptr<c10::BFloat16>();
+    c10::BFloat16* k_ptr = k.data_ptr<c10::BFloat16>();
+    c10::BFloat16* v_ptr = v.data_ptr<c10::BFloat16>();
+
+    bf16*  d_q = reinterpret_cast<bf16*>(q_ptr);
+    bf16*  d_k = reinterpret_cast<bf16*>(k_ptr);
+    bf16*  d_v = reinterpret_cast<bf16*>(v_ptr);
+    
+
+    
+    torch::Tensor l_vec = torch::empty({static_cast<const uint>(batch), 
+                                        static_cast<const uint>(qo_heads), 
+                                        static_cast<const uint>(seq_len), 
+                                        static_cast<const uint>(1)}, 
+                                        torch::TensorOptions().dtype(torch::kFloat).device(q.device()).memory_format(at::MemoryFormat::Contiguous));
+        
+
+    bf16*  o_ptr = reinterpret_cast<bf16*>(o.data_ptr<c10::BFloat16>());
+    bf16*  d_o   = reinterpret_cast<bf16*>(o_ptr);
+
+    float* l_ptr = reinterpret_cast<float*>(l_vec.data_ptr<float>());
+    float* d_l   = reinterpret_cast<float*>(l_ptr);
+
+    cudaDeviceSynchronize();
+    auto stream = at::cuda::getCurrentCUDAStream().stream(); 
+
+
+    if (head_dim == 128) {
+        using q_tile    =         st_bf<fwd_attend_ker_tile_dims<128>::qo_height, fwd_attend_ker_tile_dims<128>::tile_width>;
+        using k_tile    =         st_bf<fwd_attend_ker_tile_dims<128>::kv_height, fwd_attend_ker_tile_dims<128>::tile_width>;
+        using v_tile    =         st_bf<fwd_attend_ker_tile_dims<128>::kv_height, fwd_attend_ker_tile_dims<128>::tile_width>;
+        using l_col_vec = col_vec<st_fl<fwd_attend_ker_tile_dims<128>::qo_height, fwd_attend_ker_tile_dims<128>::tile_width>>;
+        using o_tile    =         st_bf<fwd_attend_ker_tile_dims<128>::qo_height, fwd_attend_ker_tile_dims<128>::tile_width>;
+
+        using q_global = gl<bf16,  -1, -1, -1, -1, q_tile>;
+        using k_global = gl<bf16,  -1, -1, -1, -1, k_tile>;
+        using v_global = gl<bf16,  -1, -1, -1, -1, v_tile>;
+        using l_global = gl<float, -1, -1, -1, -1, l_col_vec>;
+        using o_global = gl<bf16,  -1, -1, -1, -1, o_tile>;
+
+        using globals      = fwd_globals<128>;
+
+        q_global qg_arg{d_q, static_cast<unsigned int>(batch), static_cast<unsigned int>(qo_heads), static_cast<unsigned int>(seq_len), 128U};
+        k_global kg_arg{d_k, static_cast<unsigned int>(batch), static_cast<unsigned int>(kv_heads), static_cast<unsigned int>(seq_len), 128U};
+        v_global vg_arg{d_v, static_cast<unsigned int>(batch), static_cast<unsigned int>(kv_heads), static_cast<unsigned int>(seq_len), 128U};
+        l_global lg_arg{d_l, static_cast<unsigned int>(batch), static_cast<unsigned int>(qo_heads), 1U,   static_cast<unsigned int>(seq_len)};
+        o_global og_arg{d_o, static_cast<unsigned int>(batch), static_cast<unsigned int>(qo_heads), static_cast<unsigned int>(seq_len), 128U};
+
+        globals g{qg_arg, kg_arg, vg_arg, lg_arg, og_arg, static_cast<int>(seq_len),  static_cast<int>(text_length), static_cast<int>(hr)};
+
+        auto mem_size = kittens::MAX_SHARED_MEMORY;
+        auto threads  = NUM_WORKERS * kittens::WARP_THREADS;
+        if (has_text) {
+            // TORCH_CHECK(seq_len % (CONSUMER_WARPGROUPS*kittens::TILE_DIM*4) == 0, "sequence length must be divisible by 192");
+            dim3 grid_image(seq_len/(CONSUMER_WARPGROUPS*kittens::TILE_ROW_DIM<bf16>*4)-2, qo_heads, batch);
+            dim3 grid_text(2, qo_heads, batch);
+            if (!process_text) {
+                if (kernel_t_size == 3 && kernel_h_size == 3 && kernel_w_size == 3) {
+
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 1, 1, 1, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 1, 1, 1, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                }  else if (kernel_t_size == 3 && kernel_h_size == 3 && kernel_w_size == 5) {
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 1, 1, 2, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true,1, 1, 2, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 5 && kernel_h_size == 3 && kernel_w_size == 3) {
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 2, 1, 1, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 2, 1, 1, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                }else if (kernel_t_size ==3 && kernel_h_size == 5 && kernel_w_size == 5){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 1, 2, 2, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 1, 2, 2, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size ==5 && kernel_h_size == 6 && kernel_w_size == 1){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 2, 3, 0, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 2, 3, 0, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size ==5 && kernel_h_size == 3 && kernel_w_size == 5){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 2, 1, 2, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 2, 1, 2, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 5 && kernel_h_size == 5 && kernel_w_size == 5){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 2, 2, 2, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 2, 2, 2, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 5 && kernel_h_size == 5 && kernel_w_size == 7){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 2, 2, 3, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 2, 2, 3, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 5 && kernel_h_size == 6 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 2, 3, 5, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 2, 3, 5, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 5 && kernel_h_size == 1 && kernel_w_size == 1){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 2, 0, 0, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 2, 0, 0, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 6 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 0, 3, 5, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true, 0, 3, 5, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 5 && kernel_h_size == 1 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, true, 2, 0, 5, 5, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, true,2, 0, 5, 5, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else {
+                    // print error
+                    std::cout << "Invalid kernel size" << std::endl;
+                    //print kernel size
+                    std::cout << "Kernel size: " << kernel_t_size << " " << kernel_h_size << " " << kernel_w_size << std::endl;
+                }
+            } else {
+                cudaFuncSetAttribute(
+                    fwd_attend_ker<128, false, true, true, 1, 1, 1, 5, 6, 10>,
+                    cudaFuncAttributeMaxDynamicSharedMemorySize,
+                    mem_size
+                );
+                fwd_attend_ker<128, false, true, true, 1, 1, 1, 5, 6, 10><<<grid_text, (32*NUM_WORKERS), mem_size, stream>>>(g);
+            }
+
+        } else {
+            dim3 grid_image(seq_len/(CONSUMER_WARPGROUPS*kittens::TILE_ROW_DIM<bf16>*4), qo_heads, batch);
+            if (kernel_aspect_ratio_flag == 2){
+                if (kernel_t_size == 3 && kernel_h_size == 3 && kernel_w_size == 3) {
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 1, 1, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 1, 1, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                }  else if (kernel_t_size == 3 && kernel_h_size == 3 && kernel_w_size == 6) {
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 1, 3, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false,1, 1, 3, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 6 && kernel_h_size == 3 && kernel_w_size == 3) {
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 3, 1, 1, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 3, 1, 1, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size ==3 && kernel_h_size == 6 && kernel_w_size == 6){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 3, 3, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 3, 3, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                }else if (kernel_t_size ==3 && kernel_h_size == 6 && kernel_w_size == 3){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 3, 1, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 3, 1, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size ==6 && kernel_h_size == 3 && kernel_w_size == 6){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 3, 1, 3, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 3, 1, 3, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 6 && kernel_h_size == 6 && kernel_w_size == 6){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 3, 3, 3, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 3, 3, 3, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 6 && kernel_h_size == 1 && kernel_w_size == 1){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 3, 0, 0, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 3, 0, 0, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 6 && kernel_h_size == 1 && kernel_w_size == 6){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 3, 0, 3, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 3, 0, 3, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                }  else if (kernel_t_size == 6 && kernel_h_size == 6 && kernel_w_size == 1){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 3, 3, 0, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 3, 3, 0, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 6 && kernel_w_size == 6){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 3, 3, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 0, 3, 3, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                }  else if (kernel_t_size == 1 && kernel_h_size == 1 && kernel_w_size == 6){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 0, 3, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 0, 0, 3, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 6 && kernel_w_size == 1){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 3, 0, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 0, 3, 0, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 6 && kernel_h_size == 6 && kernel_w_size == 1){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 3, 3, 0, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 3, 3, 0, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 6 && kernel_h_size == 1 && kernel_w_size == 6){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 3, 0, 3, 6, 6, 6>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 3, 0, 3, 6, 6, 6><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else {
+                    // print error
+                    std::cout << "Invalid kernel size" << std::endl;
+                    //print kernel size
+                    std::cout << "Kernel size: " << kernel_t_size << " " << kernel_h_size << " " << kernel_w_size << std::endl;
+                }
+            }
+            else if (kernel_aspect_ratio_flag == 3) {
+                if (kernel_t_size == 3 && kernel_h_size == 3 && kernel_w_size == 3) {
+
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 1, 1, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 1, 1, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 3 && kernel_h_size == 3 && kernel_w_size == 5) {
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 1, 2, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false,1, 1, 2, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 3 && kernel_h_size == 5 && kernel_w_size == 5){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 2, 2, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 2, 2, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 3 && kernel_h_size == 6 && kernel_w_size == 1){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 3, 0, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 3, 0, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+
+                } else if (kernel_t_size == 3 && kernel_h_size == 5 && kernel_w_size == 7){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 2, 3, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 2, 3, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 3 && kernel_h_size == 5 && kernel_w_size == 9){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 2, 4, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 2, 4, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 3 && kernel_h_size == 6 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 3, 5, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 3, 5, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 3 && kernel_h_size == 6 && kernel_w_size == 3){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 3, 1, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 3, 1, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 3 && kernel_h_size == 1 && kernel_w_size == 1){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 0, 0, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 1, 0, 0, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 6 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 3, 5, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 0, 3, 5, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 5 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 2, 5, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 0, 2, 5, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 6 && kernel_w_size == 7){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 3, 3, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 0, 3, 3, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 5 && kernel_w_size == 7){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 2, 3, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 0, 2, 3, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 5 && kernel_w_size == 9){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 2, 4, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false, 0, 2, 4, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 3 && kernel_h_size == 1 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 0, 5, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false,1, 0, 5, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 3 && kernel_h_size == 3 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 1, 1, 5, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false,1, 1, 5, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 3 && kernel_w_size == 10){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 1, 5, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false,0, 1, 5, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else if (kernel_t_size == 1 && kernel_h_size == 6 && kernel_w_size == 5){
+                    cudaFuncSetAttribute(
+                        fwd_attend_ker<128, false, false, false, 0, 3, 2, 3, 6, 10>,
+                        cudaFuncAttributeMaxDynamicSharedMemorySize,
+                        mem_size
+                    );
+                    fwd_attend_ker<128, false, false, false,0, 3, 2, 3, 6, 10><<<grid_image, (32*NUM_WORKERS), mem_size, stream>>>(g);
+                } else {
+                    // print error
+                    std::cout << "Invalid kernel size" << std::endl;
+                    //print kernel size
+                    std::cout << "Kernel size: " << kernel_t_size << " " << kernel_h_size << " " << kernel_w_size << std::endl;
+                }
+            }
+
+            else {
+                std::cout << "Unsupported kernel_aspect_ratio_flag: " << kernel_aspect_ratio_flag << std::endl;
+            }
+
+        }
+        CHECK_CUDA_ERROR(cudaGetLastError());
+        cudaStreamSynchronize(stream);
+    }
+
+    return o;
+    cudaDeviceSynchronize();
+}

csrc/sliding_tile_attention/test/bench.py

@@ -0,0 +1,151 @@
+import os
+from collections import defaultdict
+
+import matplotlib.pyplot as plt
+import numpy as np
+import torch
+from st_attn import sliding_tile_attention
+
+
+def flops(batch, seqlen, nheads, headdim, causal, mode="fwd"):
+    assert mode in ["fwd", "bwd", "fwd_bwd"]
+    f = 4 * batch * seqlen**2 * nheads * headdim // (2 if causal else 1)
+    return f if mode == "fwd" else (2.5 * f if mode == "bwd" else 3.5 * f)
+
+
+def efficiency(flop, time):
+    flop = flop / 1e12
+    time = time / 1e6
+    return flop / time
+
+
+def benchmark_attention(configurations):
+    results = {'fwd': defaultdict(list), 'bwd': defaultdict(list)}
+
+    for B, H, N, D, causal in configurations:
+        print("=" * 60)
+        print(f"Timing forward and backward pass for B={B}, H={H}, N={N}, D={D}, causal={causal}")
+
+        q = torch.randn(B, H, N, D, dtype=torch.bfloat16, device='cuda', requires_grad=False).contiguous()
+        k = torch.randn(B, H, N, D, dtype=torch.bfloat16, device='cuda', requires_grad=False).contiguous()
+        v = torch.randn(B, H, N, D, dtype=torch.bfloat16, device='cuda', requires_grad=False).contiguous()
+
+        grad_output = torch.randn_like(q, requires_grad=False).contiguous()
+
+        qg = torch.zeros_like(q, requires_grad=False, dtype=torch.float).contiguous()
+        kg = torch.zeros_like(k, requires_grad=False, dtype=torch.float).contiguous()
+        vg = torch.zeros_like(v, requires_grad=False, dtype=torch.float).contiguous()
+
+        # Prepare for timing forward pass
+        start_events_fwd = [torch.cuda.Event(enable_timing=True) for _ in range(10)]
+        end_events_fwd = [torch.cuda.Event(enable_timing=True) for _ in range(10)]
+
+        torch.cuda.empty_cache()
+        torch.cuda.synchronize()
+
+        # Warmup for forward pass
+        for _ in range(10):
+            o = sliding_tile_attention(q, k, v, [[6, 6, 6]] * 24, 0, False)
+
+        # Time the forward pass
+        for i in range(10):
+            start_events_fwd[i].record()
+            o = sliding_tile_attention(q, k, v, [[6, 6, 6]] * 24, 0, False)
+            end_events_fwd[i].record()
+
+        torch.cuda.synchronize()
+        times_fwd = [s.elapsed_time(e) for s, e in zip(start_events_fwd, end_events_fwd)]
+        time_us_fwd = np.mean(times_fwd) * 1000
+
+        tflops_fwd = efficiency(flops(B, N, H, D, causal, 'fwd'), time_us_fwd)
+        results['fwd'][(D, causal)].append((N, tflops_fwd))
+
+        print(f"Average time for forward pass in us: {time_us_fwd:.2f}")
+        print(f"Average efficiency for forward pass in TFLOPS: {tflops_fwd}")
+        print("-" * 60)
+
+        # torch.cuda.empty_cache()
+        # torch.cuda.synchronize()
+
+        # # Prepare for timing backward pass
+        # start_events_bwd = [torch.cuda.Event(enable_timing=True) for _ in range(10)]
+        # end_events_bwd = [torch.cuda.Event(enable_timing=True) for _ in range(10)]
+
+        # # Warmup for backward pass
+        # for _ in range(10):
+        #     qg, kg, vg = tk.mha_backward(q, k, v, o, l_vec, grad_output, causal)
+
+        # # Time the backward pass
+        # for i in range(10):
+        #     start_events_bwd[i].record()
+        #     qg, kg, vg = tk.mha_backward(q, k, v, o, l_vec, grad_output, causal)
+        #     end_events_bwd[i].record()
+
+        # torch.cuda.synchronize()
+        # times_bwd = [s.elapsed_time(e) for s, e in zip(start_events_bwd, end_events_bwd)]
+        # time_us_bwd = np.mean(times_bwd) * 1000
+
+        # tflops_bwd = efficiency(flops(B, N, H, D, causal, 'bwd'), time_us_bwd)
+        # results['bwd'][(D, causal)].append((N, tflops_bwd))
+
+        # print(f"Average time for backward pass in us: {time_us_bwd:.2f}")
+        # print(f"Average efficiency for backward pass in TFLOPS: {tflops_bwd}")
+        print("=" * 60)
+
+        torch.cuda.empty_cache()
+        torch.cuda.synchronize()
+
+    return results
+
+
+def plot_results(results):
+    os.makedirs('benchmark_results', exist_ok=True)
+    for mode in ['fwd', 'bwd']:
+        for (D, causal), values in results[mode].items():
+            seq_lens = [x[0] for x in values]
+            tflops = [x[1] for x in values]
+
+            plt.figure(figsize=(10, 6))
+            bars = plt.bar(range(len(seq_lens)), tflops, tick_label=seq_lens)
+            plt.xlabel('Sequence Length')
+            plt.ylabel('TFLOPS')
+            plt.title(f'{mode.upper()} Pass - Head Dim: {D}, Causal: {causal}')
+            plt.grid(True)
+
+            # Adding the numerical y value on top of each bar
+            for bar in bars:
+                yval = bar.get_height()
+                plt.text(bar.get_x() + bar.get_width() / 2, yval, round(yval, 2), ha='center', va='bottom')
+
+            filename = f'benchmark_results/{mode}_D{D}_causal{causal}.png'
+            plt.savefig(filename)
+            plt.close()
+
+
+# Example list of configurations to test
+configurations = [
+    (2, 24, 82944, 128, False),
+    # (16, 16, 768*16,  128, False),
+    # (16, 16, 768*2,  128, False),
+    # (16, 16, 768*4,  128, False),
+    # (16, 16, 768*8,  128, False),
+    # (16, 16, 768*16, 128, False),
+    # (16, 16, 768,    128, True),
+    # (16, 16, 768*2,  128, True),
+    # (16, 16, 768*4,  128, True),
+    # (16, 16, 768*8,  128, True),
+    # (16, 16, 768*16, 128, True),
+    # (16, 32, 768,    64,  False),
+    # (16, 32, 768*2,  64,  False),
+    # (16, 32, 768*4,  64,  False),
+    # (16, 32, 768*8,  64,  False),
+    # (16, 32, 768*16, 64,  False),
+    # (16, 32, 768,    64,  True),
+    # (16, 32, 768*2,  64,  True),
+    # (16, 32, 768*4,  64,  True),
+    # (16, 32, 768*8,  64,  True),
+    # (16, 32, 768*16, 64,  True),
+]
+
+results = benchmark_attention(configurations)
+# plot_results(results)

csrc/sliding_tile_attention/test/test_sta.py

@@ -1,29 +1,28 @@
 import torch
-from .support_flex_sta import get_sliding_tile_attention_mask
-from fastvideo_kernel import sliding_tile_attention
+from flex_sta_ref import get_sliding_tile_attention_mask
+from st_attn import sliding_tile_attention
 from torch.nn.attention.flex_attention import flex_attention
-# from flash_attn_interface import flash_attn_func
 from tqdm import tqdm
 
 flex_attention = torch.compile(flex_attention, dynamic=False)
 
 
 def flex_test(Q, K, V, kernel_size):
-    mask = get_sliding_tile_attention_mask(kernel_size, (6, 8, 8), (18, 48, 80), 0, 'cuda', 0)
+    mask = get_sliding_tile_attention_mask(kernel_size, (6, 8, 8), (36, 48, 48), 39, 'cuda', 0)
     output = flex_attention(Q, K, V, block_mask=mask)
 
     return output
 
 
 def h100_fwd_kernel_test(Q, K, V, kernel_size):
-    o = sliding_tile_attention(Q, K, V, [kernel_size] * 24, 0, False, '18x48x80')
+    o = sliding_tile_attention(Q, K, V, [kernel_size] * 24, 39, False)
     return o
 
 
 def generate_tensor(shape, mean, std, dtype, device):
     tensor = torch.randn(shape, dtype=dtype, device=device)
 
-    magnitude = torch.norm(tensor, dim=-1, keepdim=True)
+    magnitude = torch.linalg.norm(tensor, dim=-1, keepdim=True)
     scaled_tensor = tensor * (torch.randn(magnitude.shape, dtype=dtype, device=device) * std + mean) / magnitude
 
     return scaled_tensor.contiguous()
@@ -37,7 +36,7 @@ def check_correctness(b, h, n, d, causal, mean, std, num_iterations=50, error_mo
             'max_diff': 0
         },
     }
-    kernel_size_ls = [(3, 3, 5), (3, 1, 10)]
+    kernel_size_ls = [(6, 1, 6), (6, 6, 1)]
     from tqdm import tqdm
     for kernel_size in tqdm(kernel_size_ls):
         for _ in range(num_iterations):
@@ -72,24 +71,25 @@ def check_correctness(b, h, n, d, causal, mean, std, num_iterations=50, error_mo
     return results
 
 
+def generate_error_graphs(b, h, d, causal, mean, std, error_mode='all'):
+    seq_lengths = [82944]
+
+    tk_avg_errors, tk_max_errors = [], []
+
+    for n in tqdm(seq_lengths, desc="Generating error data"):
+        results = check_correctness(b, h, n, d, causal, mean, std, error_mode=error_mode)
+
+        tk_avg_errors.append(results['TK vs FLEX']['avg_diff'])
+        tk_max_errors.append(results['TK vs FLEX']['max_diff'])
+
+
 # Example usage
-def test_sliding_tile_attention():
-    if not torch.cuda.is_available():
-        return
+b, h, d = 2, 24, 128
+causal = False
+mean = 1e-1
+std = 10
 
-    b, h, d = 2, 24, 128
-    n = 69120  # Sequence length
-    causal = False
-    mean = 1e-1
-    std = 10
-
-    # Run correctness check directly
-    results = check_correctness(b, h, n, d, causal, mean, std, error_mode='output')
-    assert results['TK vs FLEX']['avg_diff'] < 3e-6, f"Average difference: {results['TK vs FLEX']['avg_diff']} is too large"
-    assert results['TK vs FLEX']['max_diff'] < 4e-2, f"Maximum difference: {results['TK vs FLEX']['max_diff']} is too large"
-    print(f"Average difference: {results['TK vs FLEX']['avg_diff']}")
-    print(f"Maximum difference: {results['TK vs FLEX']['max_diff']}")
-
-if __name__ == "__main__":
-    test_sliding_tile_attention()
+for mode in ['output']:
+    generate_error_graphs(b, h, d, causal, mean, std, error_mode=mode)
 
+print("Error graphs generated and saved for all modes.")

docker/Dockerfile.python3.10

@@ -1,9 +1,7 @@
-FROM nvidia/cuda:12.8.0-cudnn-devel-ubuntu22.04
+FROM nvidia/cuda:12.4.1-devel-ubuntu20.04
 
 ENV DEBIAN_FRONTEND=noninteractive
 
-SHELL ["/bin/bash", "-c"]
-
 WORKDIR /FastVideo
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
@@ -11,25 +9,17 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
     git \
     ca-certificates \
     openssh-server \
-    zsh \
-    vim \
-    curl \
-    gcc-11 \
-    g++-11 \
-    clang-11 \
     && rm -rf /var/lib/apt/lists/*
 
-# Set up C++20 compilers for ThunderKittens
-RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
+RUN wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh && \
+    bash Miniconda3-latest-Linux-x86_64.sh -b -p /opt/conda && \
+    rm Miniconda3-latest-Linux-x86_64.sh
 
-# Set CUDA environment variables
-ENV CUDA_HOME=/usr/local/cuda-12.8
-ENV PATH=${CUDA_HOME}/bin:${PATH}
-ENV LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
+ENV PATH=/opt/conda/bin:$PATH
 
-# Install uv and source its environment
-RUN curl -LsSf https://astral.sh/uv/install.sh | sh && \
-    echo 'source $HOME/.local/bin/env' >> /root/.bashrc
+RUN conda create --name fastvideo-dev python=3.10.0 -y
+
+SHELL ["/bin/bash", "-c"]
 
 # Copy just the pyproject.toml first to leverage Docker cache
 COPY pyproject.toml ./
@@ -37,30 +27,22 @@ COPY pyproject.toml ./
 # Create a dummy README to satisfy the installation
 RUN echo "# Placeholder" > README.md
 
-# Create and activate virtual environment with specific Python version and seed
-RUN source $HOME/.local/bin/env && \
-    uv venv --python 3.10 --seed /opt/venv && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir --upgrade pip && \
-    uv pip install --no-cache-dir .[dev] && \
-    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.5.4/flash_attn-2.8.3%2Bcu128torch2.9-cp310-cp310-linux_x86_64.whl
+RUN conda run -n fastvideo-dev pip install --no-cache-dir --upgrade pip && \
+    conda run -n fastvideo-dev pip install --no-cache-dir .[dev] && \
+    conda run -n fastvideo-dev pip install --no-cache-dir flash-attn==2.7.4.post1 --no-build-isolation && \
+    conda clean -afy
 
 COPY . .
 
-# Install dependencies using uv and set up shell configuration
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir -e .[dev] && \
-    git config --unset-all http.https://github.com/.extraheader || true && \
-    echo 'source /opt/venv/bin/activate' >> /root/.bashrc && \
-    echo 'if [ -n "$ZSH_VERSION" ] && [ -f ~/.zshrc ]; then . ~/.zshrc; elif [ -f ~/.bashrc ]; then . ~/.bashrc; fi' > /root/.profile
+RUN conda run -n fastvideo-dev pip install --no-cache-dir -e .[dev]
 
-# Install FastVideo Unified Kernel
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    cd fastvideo-kernel && \
-    git submodule update --init --recursive && \
-    ./build.sh
+# Remove authentication headers
+RUN git config --unset-all http.https://github.com/.extraheader || true
 
+# Set up automatic conda environment activation for all shells
+RUN echo 'source /opt/conda/etc/profile.d/conda.sh' >> /root/.bashrc && \
+    echo 'conda activate fastvideo-dev' >> /root/.bashrc && \
+    # Ensure .bashrc is sourced for SSH login shells
+    echo 'if [ -f ~/.bashrc ]; then . ~/.bashrc; fi' > /root/.profile
 
-EXPOSE 22
+EXPOSE 22

docker/Dockerfile.python3.11

@@ -1,9 +1,7 @@
-FROM nvidia/cuda:12.8.0-cudnn-devel-ubuntu22.04
+FROM nvidia/cuda:12.4.1-devel-ubuntu20.04
 
 ENV DEBIAN_FRONTEND=noninteractive
 
-SHELL ["/bin/bash", "-c"]
-
 WORKDIR /FastVideo
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
@@ -11,25 +9,17 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
     git \
     ca-certificates \
     openssh-server \
-    zsh \
-    vim \
-    curl \
-    gcc-11 \
-    g++-11 \
-    clang-11 \
     && rm -rf /var/lib/apt/lists/*
 
-# Set up C++20 compilers for ThunderKittens
-RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
+RUN wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh && \
+    bash Miniconda3-latest-Linux-x86_64.sh -b -p /opt/conda && \
+    rm Miniconda3-latest-Linux-x86_64.sh
 
-# Set CUDA environment variables
-ENV CUDA_HOME=/usr/local/cuda-12.8
-ENV PATH=${CUDA_HOME}/bin:${PATH}
-ENV LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
+ENV PATH=/opt/conda/bin:$PATH
 
-# Install uv and source its environment
-RUN curl -LsSf https://astral.sh/uv/install.sh | sh && \
-    echo 'source $HOME/.local/bin/env' >> /root/.bashrc
+RUN conda create --name fastvideo-dev python=3.11.11 -y
+
+SHELL ["/bin/bash", "-c"]
 
 # Copy just the pyproject.toml first to leverage Docker cache
 COPY pyproject.toml ./
@@ -37,30 +27,22 @@ COPY pyproject.toml ./
 # Create a dummy README to satisfy the installation
 RUN echo "# Placeholder" > README.md
 
-# Create and activate virtual environment with specific Python version and seed
-RUN source $HOME/.local/bin/env && \
-    uv venv --python 3.11 --seed /opt/venv && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir --upgrade pip && \
-    uv pip install --no-cache-dir .[dev] && \
-    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.5.4/flash_attn-2.8.3%2Bcu128torch2.9-cp311-cp311-linux_x86_64.whl
+RUN conda run -n fastvideo-dev pip install --no-cache-dir --upgrade pip && \
+    conda run -n fastvideo-dev pip install --no-cache-dir .[dev] && \
+    conda run -n fastvideo-dev pip install --no-cache-dir flash-attn==2.7.4.post1 --no-build-isolation && \
+    conda clean -afy
 
 COPY . .
 
-# Install dependencies using uv and set up shell configuration
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir -e .[dev] && \
-    git config --unset-all http.https://github.com/.extraheader || true && \
-    echo 'source /opt/venv/bin/activate' >> /root/.bashrc && \
-    echo 'if [ -n "$ZSH_VERSION" ] && [ -f ~/.zshrc ]; then . ~/.zshrc; elif [ -f ~/.bashrc ]; then . ~/.bashrc; fi' > /root/.profile
+RUN conda run -n fastvideo-dev pip install --no-cache-dir -e .[dev]
 
-# Install FastVideo Unified Kernel
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    cd fastvideo-kernel && \
-    git submodule update --init --recursive && \
-    ./build.sh
+# Remove authentication headers
+RUN git config --unset-all http.https://github.com/.extraheader || true
 
+# Set up automatic conda environment activation for all shells
+RUN echo 'source /opt/conda/etc/profile.d/conda.sh' >> /root/.bashrc && \
+    echo 'conda activate fastvideo-dev' >> /root/.bashrc && \
+    # Ensure .bashrc is sourced for SSH login shells
+    echo 'if [ -f ~/.bashrc ]; then . ~/.bashrc; fi' > /root/.profile
 
-EXPOSE 22
+EXPOSE 22

docker/Dockerfile.python3.12

@@ -1,9 +1,7 @@
-FROM nvidia/cuda:12.8.0-cudnn-devel-ubuntu22.04
+FROM nvidia/cuda:12.4.1-devel-ubuntu20.04
 
 ENV DEBIAN_FRONTEND=noninteractive
 
-SHELL ["/bin/bash", "-c"]
-
 WORKDIR /FastVideo
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
@@ -11,25 +9,17 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
     git \
     ca-certificates \
     openssh-server \
-    zsh \
-    vim \
-    curl \
-    gcc-11 \
-    g++-11 \
-    clang-11 \
     && rm -rf /var/lib/apt/lists/*
 
-# Set up C++20 compilers for ThunderKittens
-RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
+RUN wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh && \
+    bash Miniconda3-latest-Linux-x86_64.sh -b -p /opt/conda && \
+    rm Miniconda3-latest-Linux-x86_64.sh
 
-# Set CUDA environment variables
-ENV CUDA_HOME=/usr/local/cuda-12.8
-ENV PATH=${CUDA_HOME}/bin:${PATH}
-ENV LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
+ENV PATH=/opt/conda/bin:$PATH
 
-# Install uv and source its environment
-RUN curl -LsSf https://astral.sh/uv/install.sh | sh && \
-    echo 'source $HOME/.local/bin/env' >> /root/.bashrc
+RUN conda create --name fastvideo-dev python=3.12.9 -y
+
+SHELL ["/bin/bash", "-c"]
 
 # Copy just the pyproject.toml first to leverage Docker cache
 COPY pyproject.toml ./
@@ -37,29 +27,22 @@ COPY pyproject.toml ./
 # Create a dummy README to satisfy the installation
 RUN echo "# Placeholder" > README.md
 
-# Create and activate virtual environment with specific Python version and seed
-RUN source $HOME/.local/bin/env && \
-    uv venv --python 3.12 --seed /opt/venv && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir --upgrade pip && \
-    uv pip install --no-cache-dir .[dev] && \
-    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.5.4/flash_attn-2.8.3%2Bcu128torch2.9-cp312-cp312-linux_x86_64.whl
+RUN conda run -n fastvideo-dev pip install --no-cache-dir --upgrade pip && \
+    conda run -n fastvideo-dev pip install --no-cache-dir .[dev] && \
+    conda run -n fastvideo-dev pip install --no-cache-dir flash-attn==2.7.4.post1 --no-build-isolation && \
+    conda clean -afy
 
 COPY . .
 
-# Install dependencies using uv and set up shell configuration
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir -e .[dev] && \
-    git config --unset-all http.https://github.com/.extraheader || true && \
-    echo 'source /opt/venv/bin/activate' >> /root/.bashrc && \
-    echo 'if [ -n "$ZSH_VERSION" ] && [ -f ~/.zshrc ]; then . ~/.zshrc; elif [ -f ~/.bashrc ]; then . ~/.bashrc; fi' > /root/.profile
+RUN conda run -n fastvideo-dev pip install --no-cache-dir -e .[dev]
 
-# Install FastVideo Unified Kernel
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    cd fastvideo-kernel && \
-    git submodule update --init --recursive && \
-    ./build.sh
+# Remove authentication headers
+RUN git config --unset-all http.https://github.com/.extraheader || true
 
-EXPOSE 22
+# Set up automatic conda environment activation for all shells
+RUN echo 'source /opt/conda/etc/profile.d/conda.sh' >> /root/.bashrc && \
+    echo 'conda activate fastvideo-dev' >> /root/.bashrc && \
+    # Ensure .bashrc is sourced for SSH login shells
+    echo 'if [ -f ~/.bashrc ]; then . ~/.bashrc; fi' > /root/.profile
+
+EXPOSE 22

docker/Dockerfile.python3.12.cuda12.9.1

@@ -1,66 +0,0 @@
-FROM nvidia/cuda:12.9.1-cudnn-devel-ubuntu22.04
-
-ENV DEBIAN_FRONTEND=noninteractive
-
-SHELL ["/bin/bash", "-c"]
-
-WORKDIR /FastVideo
-
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    wget \
-    git \
-    ca-certificates \
-    openssh-server \
-    zsh \
-    vim \
-    curl \
-    gcc-11 \
-    g++-11 \
-    clang-11 \
-    && rm -rf /var/lib/apt/lists/*
-
-# Set up C++20 compilers for ThunderKittens
-RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
-
-# Set CUDA environment variables
-ENV CUDA_HOME=/usr/local/cuda-12.9
-ENV PATH=${CUDA_HOME}/bin:${PATH}
-ENV LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
-
-# Install uv and source its environment
-RUN curl -LsSf https://astral.sh/uv/install.sh | sh && \
-    echo 'source $HOME/.local/bin/env' >> /root/.bashrc
-
-# Copy just the pyproject.toml first to leverage Docker cache
-COPY pyproject.toml ./
-
-# Create a dummy README to satisfy the installation
-RUN echo "# Placeholder" > README.md
-
-# Create and activate virtual environment with specific Python version and seed
-RUN source $HOME/.local/bin/env && \
-    uv venv --python 3.12 --seed /opt/venv && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir --upgrade pip && \
-    uv pip install --no-cache-dir .[dev] && \
-    uv pip install --no-cache-dir flash-attn==2.8.3 --no-build-isolation
-
-COPY . .
-
-# Install dependencies using uv and set up shell configuration
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir -e .[dev] && \
-    git config --unset-all http.https://github.com/.extraheader || true && \
-    echo 'source /opt/venv/bin/activate' >> /root/.bashrc && \
-    echo 'if [ -n "$ZSH_VERSION" ] && [ -f ~/.zshrc ]; then . ~/.zshrc; elif [ -f ~/.bashrc ]; then . ~/.bashrc; fi' > /root/.profile
-
-# Install FastVideo Unified Kernel
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    cd fastvideo-kernel && \
-    git submodule update --init --recursive && \
-    ./build.sh
-
-
-EXPOSE 22

docker/rocm7_1.Dockerfile

@@ -1,58 +0,0 @@
-FROM rocm/pytorch:rocm7.1_ubuntu22.04_py3.10_pytorch_release_2.9.1
-
-ENV DEBIAN_FRONTEND=noninteractive
-
-SHELL ["/bin/bash", "-c"]
-
-WORKDIR /FastVideo
-
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    wget \
-    git \
-    ca-certificates \
-    openssh-server \
-    zsh \
-    vim \
-    curl \
-    gcc-11 \
-    g++-11 \
-    clang-11 \
-    && rm -rf /var/lib/apt/lists/*
-
-# Set up C++20 compilers for ThunderKittens
-RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
-
-# Install uv and source its environment
-RUN curl -LsSf https://astral.sh/uv/install.sh | sh && \
-    echo 'source $HOME/.local/bin/env' >> /root/.bashrc
-
-# Copy just the pyproject.toml first to leverage Docker cache
-COPY pyproject_other.toml ./pyproject.toml
-
-# Create a dummy README to satisfy the installation
-RUN echo "# Placeholder" > README.md
-
-# Create and activate virtual environment with specific Python version and seed
-RUN source $HOME/.local/bin/env && \
-    uv venv --python 3.10 --seed /opt/venv && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir --upgrade pip
-
-COPY . .
-
-# Install dependencies using uv and set up shell configuration
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    uv pip install --no-cache-dir -e .[rocm] && \
-    git config --unset-all http.https://github.com/.extraheader || true && \
-    echo 'source /opt/venv/bin/activate' >> /root/.bashrc && \
-    echo 'if [ -n "$ZSH_VERSION" ] && [ -f ~/.zshrc ]; then . ~/.zshrc; elif [ -f ~/.bashrc ]; then . ~/.bashrc; fi' > /root/.profile
-
-# Install FastVideo Unified Kernel
-RUN source $HOME/.local/bin/env && \
-    source /opt/venv/bin/activate && \
-    cd fastvideo-kernel && \
-    git submodule update --init --recursive && \
-    ./build.sh --rocm
-
-EXPOSE 22

docs/Makefile

@@ -0,0 +1,25 @@
+# Minimal makefile for Sphinx documentation
+#
+
+# You can set these variables from the command line, and also
+# from the environment for the first two.
+SPHINXOPTS    ?=
+SPHINXBUILD   ?= sphinx-build
+SOURCEDIR     = source
+BUILDDIR      = build
+
+# Put it first so that "make" without argument is like "make help".
+help:
+	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+
+.PHONY: help Makefile
+
+# Catch-all target: route all unknown targets to Sphinx using the new
+# "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
+%: Makefile
+	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+
+clean:
+	@$(SPHINXBUILD) -M clean "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+	rm -rf "$(SOURCEDIR)/getting_started/examples"
+	rm -rf "$(SOURCEDIR)/inference/examples"

docs/README.md

@@ -1,39 +1,20 @@
-# FastVideo Documentation
+# FastVideo documents
 
-This directory contains the FastVideo documentation built with MkDocs.
-
-## Build the docs locally
+## Build the docs
 
 ```bash
-# Install dependencies
-pip install -r requirements-mkdocs.txt
+# Install dependencies.
+pip install -r requirements-docs.txt
 
-# Serve docs with live reload (recommended for development)
-mkdocs serve
-
-# Or build static site
-mkdocs build
+# Build the docs.
+make clean
+make html
 ```
 
-## View the docs
-
-### Development server (with live reload)
+## Open the docs with your browser
 
 ```bash
-mkdocs serve
+python -m http.server -d build/html/
 ```
 
-Then open your browser to: http://127.0.0.1:8000
-
-### Static build
-
-```bash
-mkdocs build
-python -m http.server -d site/
-```
-
-Then open your browser to: http://localhost:8000
-
-## Automatic Deployment
-
-Documentation is automatically built and deployed to GitHub Pages when changes are pushed to the `main` branch via the `.github/workflows/docs.yml` workflow.
+Launch your browser and open localhost:8000.

docs/api/fastvideo.md

@@ -1,248 +0,0 @@
-# FastVideo API Reference
-
-This page contains the complete API reference for the FastVideo library.
-
-## fastvideo
-
-### Modules
-
-| Name | Description |
-|------|-------------|
-| [attention](#fastvideoattention) | Attention mechanisms and backends for video generation |
-| [configs](#fastvideoconfigs) | Configuration classes for pipelines, models, and sampling |
-| [distributed](#fastvideodistributed) | Distributed execution and communication utilities |
-| [entrypoints](#fastvideoentrypoints) | Main API entry points for video generation |
-| [models](#fastvideomodels) | Model implementations (transformers, VAEs, schedulers) |
-| [pipelines](#fastvideopipelines) | Core pipeline classes for video diffusion |
-| [training](#fastvideotraining) | Training utilities and helpers |
-| [workflow](#fastvideoworkflow) | Workflow management and orchestration |
-| [dataset](#fastvideodataset) | Dataset handling and preprocessing |
-| [layers](#fastvideolayers) | Custom neural network layers |
-| [platforms](#fastvideoplatforms) | Platform-specific implementations |
-| [utils](#fastvideoutils) | Utility functions and helpers |
-| [worker](#fastvideoworker) | Execution workers for video generation |
-
-## fastvideo.attention
-
-::: fastvideo.attention
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-## fastvideo.configs
-
-::: fastvideo.configs
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-### Submodules
-
-#### fastvideo.configs.pipelines
-
-::: fastvideo.configs.pipelines
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-#### fastvideo.configs.models
-
-::: fastvideo.configs.models
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-#### fastvideo.configs.sample
-
-::: fastvideo.configs.sample
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-## fastvideo.distributed
-
-::: fastvideo.distributed
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-## fastvideo.entrypoints
-
-::: fastvideo.entrypoints
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-## fastvideo.models
-
-::: fastvideo.models
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-### Submodules
-
-#### fastvideo.models.registry
-
-::: fastvideo.models.registry
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-#### fastvideo.models.loader
-
-::: fastvideo.models.loader
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-## fastvideo.pipelines
-
-::: fastvideo.pipelines
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-### Submodules
-
-#### fastvideo.pipelines.composed_pipeline_base
-
-::: fastvideo.pipelines.composed_pipeline_base
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-#### fastvideo.pipelines.lora_pipeline
-
-::: fastvideo.pipelines.lora_pipeline
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-#### fastvideo.pipelines.pipeline_batch_info
-
-::: fastvideo.pipelines.pipeline_batch_info
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-#### fastvideo.pipelines.pipeline_registry
-
-::: fastvideo.pipelines.pipeline_registry
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-#### fastvideo.pipelines.stages
-
-::: fastvideo.pipelines.stages
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 4
-
-## fastvideo.training
-
-::: fastvideo.training
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-## fastvideo.workflow
-
-::: fastvideo.workflow
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-## fastvideo.dataset
-
-::: fastvideo.dataset
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-## fastvideo.layers
-
-::: fastvideo.layers
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-## fastvideo.platforms
-
-::: fastvideo.platforms
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3
-
-## fastvideo.utils
-
-::: fastvideo.utils
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 3
-
-## fastvideo.worker
-
-::: fastvideo.worker
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      show_submodules: true
-      heading_level: 3

docs/api/summary.md

@@ -1,27 +0,0 @@
-# API Summary
-
-This page provides a quick overview of the main FastVideo API components.
-
-## Video Generator
-
-::: fastvideo.VideoGenerator
-    options:
-      show_root_heading: false
-      show_source: false
-      heading_level: 3
-
-## Initialization Configuration
-
-::: fastvideo.PipelineConfig
-    options:
-      show_root_heading: false
-      show_source: false
-      heading_level: 3
-
-## Sampling Configuration
-
-::: fastvideo.SamplingParam
-    options:
-      show_root_heading: false
-      show_source: false
-      heading_level: 3

docs/assets/custom.css

@@ -1,43 +0,0 @@
-.vertical-table-header th.head:not(.stub) {
-    writing-mode: sideways-lr;
-    white-space: nowrap;
-    max-width: 0;
-}
-
-/* Keep header cell paragraph content tight (avoid CSS nesting for compatibility) */
-.vertical-table-header th.head:not(.stub) p {
-    margin: 0;
-}
-
-/* Image sizing classes */
-.image-small {
-    max-width: 200px;
-    height: auto;
-}
-
-.image-medium {
-    max-width: 400px;
-    height: auto;
-}
-
-.image-large {
-    max-width: 600px;
-    height: auto;
-}
-
-.image-full {
-    max-width: 100%;
-    height: auto;
-}
-
-/* Responsive images */
-img {
-    max-width: 100%;
-    height: auto;
-}
-
-/* Center images */
-.image-center {
-    display: block;
-    margin: 0 auto;
-}

docs/assets/logos/icon_simple.svg

@@ -1,6 +0,0 @@
-<svg width="160" height="93" viewBox="0 0 160 93" fill="none" xmlns="http://www.w3.org/2000/svg">
-<path d="M28.8511 91.66L57.6319 1.86368H64.5394L35.7585 91.66H28.8511Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-<path d="M15.0376 91.66L43.8185 1.86368H46.1209L17.3401 91.66H15.0376Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-<path d="M1.22217 91.66L30.003 1.86366H31.1543L2.3734 91.66H1.22217Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.15122"/>
-<path d="M71.4465 1.86483L42.666 91.6599H69.144L78.3538 58.2746H123.251L129.007 39.855H84.1099L89.866 22.5868H152.032L157.788 1.86483H71.4465Z" fill="#356CFF" stroke="#356CFF" stroke-width="2.30244"/>
-</svg>

docs/assets/logos/logo.svg

@@ -1,18 +0,0 @@
-<svg width="252" height="105" viewBox="0 0 252 105" fill="none" xmlns="http://www.w3.org/2000/svg">
-<path d="M89.4843 55.5457H101.361L87.7028 101H74.638L89.4843 55.5457Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M96.0167 1.00057H112.645L118.583 48.273H104.924L103.737 39.7882H79.9827L67.5117 55.5457H85.3273L43.1638 101H28.3174L22.3789 55.5457H33.6621L38.4129 91.3031L58.604 68.2729H44.3515L96.0167 1.00057ZM100.768 13.1217L87.7028 29.4852H103.143L100.768 13.1217Z" fill="#356CFF"/>
-<path d="M37.2252 1.00057L22.3789 48.273H36.0375L40.7884 30.6974L62.6727 30.6974L69.6727 21.0005L43.7576 21.0004L46.7269 11.9096L77.6727 11.9096L86 1.00057L37.2252 1.00057Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M108.488 55.5457L94.2351 101C94.2351 101 105.518 101 120.959 101C136.399 101 144.078 93.0133 148.276 79.788C152.432 68.0157 153.027 55.5457 136.399 55.5457C119.771 55.5457 108.488 55.5457 108.488 55.5457ZM109.081 90.697L116.802 65.8487C116.802 65.8487 120.959 65.8487 132.242 65.8487C143.525 65.8487 137.586 78.5759 135.211 84.0304C133.307 88.4021 127.491 90.697 122.74 90.697C117.989 90.697 109.081 90.697 109.081 90.697Z" fill="#356CFF"/>
-<path d="M173.188 1.00056L168.625 11.9096C168.625 11.9096 149.386 11.9092 142.525 11.9095C135.664 11.9098 136.586 20.3944 141.337 20.3944H159.747C168.654 20.3944 166.961 33.6899 163.904 38.5761C160.467 44.0675 157.371 48.273 148.463 48.273L125.188 48.273L124 37.97L147.87 37.97C153.808 37.97 156.184 29.4852 151.433 29.4852H131.836C120.142 29.4852 125.897 1.00043 141.337 1.00043L173.188 1.00056Z" fill="#356CFF"/>
-<path d="M179.938 1.00056L175.688 11.9096L191.221 11.9096L179.938 48.273H192.409L203.692 11.9096L219.132 11.9095L223.289 1.00043L179.938 1.00056Z" fill="#356CFF"/>
-<path d="M161.341 55.5457H202.845L198.5 65.8487H169.654L167.279 73.7268H188.5L184.749 82.8177H164.31L161.934 90.697H190.251L186.624 101H146.494L161.341 55.5457Z" fill="#356CFF"/>
-<path fill-rule="evenodd" clip-rule="evenodd" d="M230.821 54.9391C255.169 54.9391 251.776 67.0602 249.231 77.9692C246.686 88.8783 240.917 101 217.757 101C194.596 101 195.606 88.8783 199.347 77.9692C203.089 67.0602 206.473 54.9391 230.821 54.9391ZM237.948 77.9692C239.984 70.6965 240.917 65.242 228.446 65.242C215.975 65.242 211.818 71.9087 210.037 77.9692C208.255 84.0298 208.255 91.3025 219.538 91.3025C230.821 91.3025 235.911 85.2419 237.948 77.9692Z" fill="#356CFF"/>
-<path d="M173.188 1.00056L168.625 11.9096C168.625 11.9096 149.386 11.9092 142.525 11.9095C135.664 11.9098 136.586 20.3944 141.337 20.3944M173.188 1.00056C173.188 1.00056 156.777 1.00043 141.337 1.00043M173.188 1.00056L141.337 1.00043M141.337 20.3944C146.088 20.3944 150.839 20.3944 159.747 20.3944M141.337 20.3944H159.747M159.747 20.3944C168.654 20.3944 166.961 33.6899 163.904 38.5761C160.467 44.0675 157.371 48.273 148.463 48.273M148.463 48.273C139.556 48.273 125.188 48.273 125.188 48.273M148.463 48.273L125.188 48.273M125.188 48.273L124 37.97M124 37.97C124 37.97 141.931 37.97 147.87 37.97M124 37.97L147.87 37.97M147.87 37.97C153.808 37.97 156.184 29.4852 151.433 29.4852M151.433 29.4852C146.682 29.4852 138.962 29.4852 131.836 29.4852M151.433 29.4852H131.836M131.836 29.4852C120.142 29.4852 125.897 1.00043 141.337 1.00043M37.2252 1.00057L22.3789 48.273H36.0375L40.7884 30.6974L62.6727 30.6974L69.6727 21.0005L43.7576 21.0004L46.7269 11.9096L77.6727 11.9096L86 1.00057L37.2252 1.00057ZM96.0167 1.00057H112.645L118.583 48.273H104.924L103.737 39.7882H79.9827L67.5117 55.5457H85.3273L43.1638 101H28.3174L22.3789 55.5457H33.6621L38.4129 91.3031L58.604 68.2729H44.3515L96.0167 1.00057ZM87.7028 29.4852L100.768 13.1217L103.143 29.4852H87.7028ZM89.4843 55.5457H101.361L87.7028 101H74.638L89.4843 55.5457ZM108.488 55.5457L94.2351 101C94.2351 101 105.518 101 120.959 101C136.399 101 144.078 93.0133 148.276 79.788C152.432 68.0157 153.027 55.5457 136.399 55.5457C119.771 55.5457 108.488 55.5457 108.488 55.5457ZM116.802 65.8487L109.081 90.697C109.081 90.697 117.989 90.697 122.74 90.697C127.491 90.697 133.307 88.4021 135.211 84.0304C137.586 78.5759 143.525 65.8487 132.242 65.8487C120.959 65.8487 116.802 65.8487 116.802 65.8487ZM179.938 1.00056L175.688 11.9096L191.221 11.9096L179.938 48.273H192.409L203.692 11.9096L219.132 11.9095L223.289 1.00043L179.938 1.00056ZM161.341 55.5457H202.845L198.5 65.8487H169.654L167.279 73.7268H188.5L184.749 82.8177H164.31L161.934 90.697H190.251L186.624 101H146.494L161.341 55.5457ZM230.821 54.9391C255.169 54.9391 251.776 67.0602 249.231 77.9692C246.686 88.8783 240.917 101 217.757 101C194.596 101 195.606 88.8783 199.347 77.9692C203.089 67.0602 206.473 54.9391 230.821 54.9391ZM228.446 65.242C240.917 65.242 239.984 70.6965 237.948 77.9692C235.911 85.2419 230.821 91.3025 219.538 91.3025C208.255 91.3025 208.255 84.0298 210.037 77.9692C211.818 71.9087 215.975 65.242 228.446 65.242Z" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M15.2524 55.5451L21.191 100.999L24.7541 100.999L18.8156 55.5451L15.2524 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M8.12646 55.5451L14.065 100.999L15.2527 100.999L9.31417 55.5451L8.12646 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M1 55.5451L6.93853 100.999L7.53239 100.999L1.59385 55.5451L1 55.5451Z" fill="#356CFF" stroke="#356CFF" stroke-width="0.593853"/>
-<path d="M15.2524 48.2724L30.0988 1H33.6619L18.8156 48.2724H15.2524Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M8.12646 48.2724L22.9728 1H24.1605L9.31417 48.2724H8.12646Z" fill="#356CFF" stroke="#356CFF" stroke-width="1.18771"/>
-<path d="M1 48.2724L15.8463 1H16.4402L1.59385 48.2724H1Z" fill="#356CFF" stroke="#356CFF" stroke-width="0.593853"/>
-<path d="M85.3271 55.5457H67.5116L87 12.7363L44.3513 68.2729H58.6038L43.1636 101L85.3271 55.5457Z" fill="#FDC717" stroke="#FDC717" stroke-width="1.18771" stroke-miterlimit="16"/>
-</svg>

docs/attention/developer/index.md

@@ -1,186 +0,0 @@
-# Adding a New Attention Backend
-
-FastVideo allows integrating new attention mechanisms easily. This guide walks you through adding a new backend (e.g., `MyNewAttn`).
-
-## 1. Implement the Backend (Python)
-
-Create a new file in `fastvideo/attention/backends/` (e.g., `mynew_attn.py`).
-
-Your implementation should inherit from `AttentionBackend` defined in `abstract.py`.
-
-```python
-# fastvideo/attention/backends/mynew_attn.py
-import torch
-from .abstract import AttentionBackend
-# Import the context manager to access metadata (optional)
-from fastvideo.forward_context import get_forward_context
-
-# Import compiled kernel if applicable (see Section 2)
-try:
-    # Import from the top-level package
-    from fastvideo_kernel import my_compiled_attn_func
-except ImportError:
-    my_compiled_attn_func = None
-
-class MyNewAttnBackend(AttentionBackend):
-    def process_inputs(self, q, k, v, **kwargs):
-        # Pre-process inputs if necessary
-        return q, k, v
-
-    def forward(self, q, k, v, **kwargs):
-        # Optional: Access extra metadata passed via ForwardContext
-        # Only needed if your backend requires global state (e.g. window_size)
-        try:
-            context = get_forward_context()
-            metadata = context.attn_metadata
-            # Example: window_size = metadata.window_size
-        except (AssertionError, AttributeError):
-            # Handle case where context is not set (e.g. standard inference)
-            pass
-
-        if my_compiled_attn_func is not None:
-            return my_compiled_attn_func(q, k, v)
-        else:
-            # Fallback implementation (e.g., Triton or pure PyTorch)
-            return self.fallback_impl(q, k, v)
-```
-
-## 2. Passing Extra Information via ForwardContext (Optional)
-
-FastVideo uses a `ForwardContext` to pass global metadata (like current timestep, batch info, or custom attention configurations) to attention backends without changing the `forward` signature of every layer. **This is optional and only required if your backend needs dynamic per-step information.**
-
-To use this:
-
-1. **Set Context**: In your pipeline or generation loop, use the `set_forward_context` context manager.
-2. **Access Context**: Inside your attention backend, use `get_forward_context()`.
-
-See [`docs/attention/sta/index.md`](../sta/index.md) (Sliding Tile Attention) for an example of how complex configuration (window sizes) is passed this way.
-
-## 3. Adding Compiled Kernels (C++/CUDA)
-
-If your backend requires custom CUDA kernels, you need to add them to the `fastvideo-kernel` package.
-
-### A. Add Source Files
-
-Place your kernel implementation files in `fastvideo-kernel/csrc/attention/`.
-
-* `mynew_attn.cu` (CUDA implementation)
-* `mynew_attn.h` (Optional headers)
-
-### B. Register in Extension
-
-Update `fastvideo-kernel/csrc/common_extension.cpp` to expose your function to Python.
-
-```cpp
-// 1. Declare external function
-#ifdef COMPILE_MYNEW_ATTN
-extern torch::Tensor mynew_attn_forward(torch::Tensor q, torch::Tensor k, torch::Tensor v);
-#endif
-
-// 2. Register in module
-PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
-    // ... other kernels ...
-
-#ifdef COMPILE_MYNEW_ATTN
-    m.def("mynew_attn_fwd", torch::wrap_pybind_function(mynew_attn_forward), "My New Attention Forward");
-#endif
-}
-```
-
-### C. Update CMakeLists.txt
-
-Update `fastvideo-kernel/CMakeLists.txt` to compile your new files.
-
-**Case 1: General CUDA Kernel (Runs on all GPUs)**
-Add your source file directly to `EXTENSION_SOURCES` and define the compilation flag.
-
-```cmake
-# Add to EXTENSION_SOURCES
-list(APPEND EXTENSION_SOURCES csrc/attention/mynew_attn.cu)
-
-# Add compilation definition for common_extension.cpp
-list(APPEND COMPILE_DEFS COMPILE_MYNEW_ATTN)
-```
-
-**Case 2: ThunderKittens Kernel (Hopper H100 Only)**
-If your kernel uses ThunderKittens (TK), it requires specific architecture flags (`sm_90a`). Add it inside the `ENABLE_TK_KERNELS` block.
-
-```cmake
-if(ENABLE_TK_KERNELS)
-    # Add source only if TK is enabled
-    list(APPEND EXTENSION_SOURCES csrc/attention/mynew_attn_tk.cu)
-    
-    # Add definition to guard registration
-    list(APPEND COMPILE_DEFS TK_COMPILE_MYNEW_ATTN)
-endif()
-```
-
-### D. Expose in Python Ops
-
-Update `fastvideo-kernel/python/fastvideo_kernel/ops.py` to make the function importable and handle fallbacks gracefully.
-
-```python
-# fastvideo-kernel/python/fastvideo_kernel/ops.py
-
-# Try to load C++ extension symbols
-try:
-    from fastvideo_kernel._C import fastvideo_kernel_ops
-    mynew_attn_fwd = getattr(fastvideo_kernel_ops, "mynew_attn_fwd", None)
-except ImportError:
-    mynew_attn_fwd = None
-
-def my_compiled_attn_func(q, k, v):
-    # Runtime check: use C++ kernel if available, else fallback
-    if mynew_attn_fwd is not None:
-        return mynew_attn_fwd(q, k, v)
-    else:
-        # Call Triton/Python fallback
-        return mynew_attn_triton(q, k, v)
-```
-
-### E. Expose in Package Init
-
-Update `fastvideo-kernel/python/fastvideo_kernel/__init__.py` to export the function.
-
-```python
-from fastvideo_kernel.ops import (
-    my_compiled_attn_func,
-    # ...
-)
-
-__all__ = [
-    "my_compiled_attn_func",
-    # ...
-]
-```
-
-## 4. Register the Backend
-
-Update `fastvideo/attention/backends/__init__.py` to export your new class.
-
-```python
-from .mynew_attn import MyNewAttnBackend
-```
-
-## 5. Platform Integration
-
-If your backend requires specific platform checks (e.g., checking for H100 support), handle that in `fastvideo/platforms/cuda.py` or within your backend's `__init__`.
-
-## 6. Add Documentation
-
-Create a new documentation page for your backend to explain its usage, installation (if custom kernels are needed), and features.
-
-1. **Create Directory**: `docs/attention/mynew_attn/`
-2. **Create Index**: `docs/attention/mynew_attn/index.md`
-3. **Update Navigation**: Add an entry to `mkdocs.yml` under the "Attention" tab.
-
-## Checklist
-
-* [ ] Created `fastvideo/attention/backends/mynew_attn.py`.
-* [ ] (Optional) Added CUDA kernels in `fastvideo-kernel/csrc/attention/`.
-* [ ] (Optional) Updated `common_extension.cpp` and `CMakeLists.txt`.
-* [ ] (Optional) Exposed kernel in `fastvideo-kernel/python/fastvideo_kernel/ops.py`.
-* [ ] (Optional) Exported kernel in `fastvideo-kernel/python/fastvideo_kernel/__init__.py`.
-* [ ] Implemented `forward` method respecting the standard signature.
-* [ ] Added unit tests in `tests/`.
-* [ ] Added documentation in `docs/attention/` and updated `mkdocs.yml`.

docs/attention/index.md

@@ -1,53 +0,0 @@
-# FastVideo Attention Kernels
-
-FastVideo provides highly optimized custom attention kernels to accelerate video generation.
-
-## Supported Kernels
-
-* **[Video Sparse Attention (VSA)](vsa/index.md)**: Sparse attention mechanism selecting top-k blocks.
-* **[Sliding Tile Attention (STA)](sta/index.md)**: Optimized attention for window-based video generation.
-
-## General Build Instructions
-
-These instructions apply to building the `fastvideo-kernel` package from source, which includes both STA and VSA kernels.
-
-### Prerequisites
-
-* **PyTorch**: 2.5.0+
-* **CUDA**: 12.4+ (12.8 recommended for best performance)
-* **C++ Compiler**: GCC 11+ (C++20 support required for ThunderKittens)
-
-Install system dependencies:
-
-```bash
-sudo apt update
-sudo apt install -y gcc-11 g++-11 clang-11 ninja-build
-
-# Set gcc-11 as default
-sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
-```
-
-Set up your CUDA environment variables (adjust version as needed):
-
-```bash
-export CUDA_HOME=/usr/local/cuda-12.8
-export PATH=${CUDA_HOME}/bin:${PATH}
-export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
-```
-
-### Compile and Install
-
-Clone the repository and build the kernel:
-
-```bash
-# Clone recursively to get ThunderKittens submodule
-git clone --recursive https://github.com/hao-ai-lab/FastVideo.git
-cd FastVideo/fastvideo-kernel
-
-# Build and install
-./build.sh
-```
-
-The build script automatically detects your GPU architecture:
-* **H100 (sm_90a)**: Compiles optimized C++ ThunderKittens kernels.
-* **Other (A100, etc.)**: Skips C++ compilation; installs Python package with Triton kernels.

docs/attention/sta/index.md

@@ -1,36 +0,0 @@
-# Sliding Tile Attention (STA)
-
-Optimized attention for window-based video generation (e.g., HunyuanVideo).
-
-## Installation
-
-STA is included in the `fastvideo-kernel` package. See the [main Attention page](../index.md) for build instructions.
-
-## Usage
-
-```python
-from fastvideo_kernel import sliding_tile_attention
-
-# q, k, v: [batch_size, num_heads, seq_length, head_dim]
-# window_size: List of (t, h, w) tiles. Tile size is (6, 8, 8).
-# text_length: Number of text tokens (0-256)
-
-out = sliding_tile_attention(
-    q, k, v, 
-    window_size=[(3, 3, 3)],  # Example window
-    text_length=256
-)
-```
-
-## Citation
-
-If you use Sliding Tile Attention in your research, please cite:
-
-```bibtex
-@article{zhang2025fast,
-  title={Fast video generation with sliding tile attention},
-  author={Zhang, Peiyuan and Chen, Yongqi and Su, Runlong and Ding, Hangliang and Stoica, Ion and Liu, Zhengzhong and Zhang, Hao},
-  journal={arXiv preprint arXiv:2502.04507},
-  year={2025}
-}
-```

docs/attention/vsa/index.md

@@ -1,36 +0,0 @@
-# Video Sparse Attention (VSA)
-
-Sparse attention mechanism selecting top-k blocks.
-
-## Installation
-
-VSA is included in the `fastvideo-kernel` package. See the [main Attention page](../index.md) for build instructions.
-
-## Usage
-
-```python
-from fastvideo_kernel import video_sparse_attn
-
-# q, k, v: [batch_size, num_heads, seq_len, head_dim]
-# variable_block_sizes: Number of valid tokens per block
-# topk: Number of blocks to attend
-
-output = video_sparse_attn(
-    q, k, v, 
-    variable_block_sizes=block_sizes, 
-    topk=32
-)
-```
-
-## Citation
-
-If you use Video Sparse Attention in your research, please cite:
-
-```bibtex
-@article{zhang2025vsa,
-  title={Vsa: Faster video diffusion with trainable sparse attention},
-  author={Zhang, Peiyuan and Chen, Yongqi and Huang, Haofeng and Lin, Will and Liu, Zhengzhong and Stoica, Ion and Xing, Eric and Zhang, Hao},
-  journal={arXiv preprint arXiv:2505.13389},
-  year={2025}
-}
-```

docs/contributing/profiling.md

@@ -1,53 +0,0 @@
-# Profiling FastVideo
-
-!!! warning
-    Profiling is only intended for FastVideo developers and maintainers to understand the proportion of time spent in different parts of the codebase. **FastVideo end-users should never turn on profiling** as it will significantly slow down inference.
-
-## Profiling with PyTorch
-
-FastVideo exposes a process-wide torch profiler that you can enable via environment variables. Set `FASTVIDEO_TORCH_PROFILER_DIR` to an absolute directory path to start collecting traces, and specify the regions you want recorded with `FASTVIDEO_TORCH_PROFILE_REGIONS`:
-
-```bash
-FASTVIDEO_TORCH_PROFILER_DIR=/mnt/traces/fastvideo \
-FASTVIDEO_TORCH_PROFILE_REGIONS="profiler_region_model_loading,profiler_region_training_step"
-```
-
-All profiled regions must be registered in `fastvideo.profiler`; the current list includes:
-
-- `profiler_region_model_loading` — pipeline/module loading
-- `profiler_region_inference_pre_denoising`
-- `profiler_region_inference_denoising`
-- `profiler_region_inference_post_denoising`
-- `profiler_region_training_checkpoint_saving`
-- `profiler_region_training_dit`
-- `profiler_region_training_validation`
-- `profiler_region_training_epoch`
-- `profiler_region_training_step`
-- `profiler_region_training_backward`
-- `profiler_region_training_optimizer`
-- `profiler_region_distillation_teacher_forward`
-- `profiler_region_distillation_student_forward`
-- `profiler_region_distillation_loss`
-- `profiler_region_distillation_update`
-
-While profiling is enabled, FastVideo records additional annotations:
-
-- `fastvideo.region::<name>` spans are emitted when entering a region.
-- `fastvideo.profiler.enable_collection` / `fastvideo.profiler.disable_collection` events mark when torch profiler collection is toggled on or off.
-
-Only one profiler instance is created per process; subsequent pipelines reuse the same controller. If you set `FASTVIDEO_TORCH_PROFILE_REGIONS` incorrectly (e.g. misspelled name), FastVideo logs a warning and ignores that entry.
-
-Additional knobs:
-
-- `FASTVIDEO_TORCH_PROFILER_RECORD_SHAPES`
-- `FASTVIDEO_TORCH_PROFILER_WITH_PROFILE_MEMORY`
-- `FASTVIDEO_TORCH_PROFILER_WITH_STACK`
-- `FASTVIDEO_TORCH_PROFILER_WITH_FLOPS`
-
-Traces can be visualized using <https://ui.perfetto.dev/>.
-
-### Best Practices
-
-- Keep the profiled step count small; traces can be large and slow down job shutdown while the profiler flushes data.
-- After profiling, clean up trace directories to avoid filling disk storage.
-- When adding new regions, register them in `fastvideo.profiler` and wrap the corresponding code block with `with self.profiler_controller.region("your_region"):` or the `@profile_region` decorator.

docs/contributing/testing.md

@@ -1,131 +0,0 @@
-# Testing in FastVideo
-
-This guide explains how to add and run tests in FastVideo. The testing suite is divided into several categories to ensure correctness across components, training workflows, and inference quality.
-
-## Test Types
-
-* **Unit Tests**: Located in `fastvideo/tests/dataset`, `fastvideo/tests/entrypoints`, and `fastvideo/tests/workflow`. These test individual functions and classes.
-* **Component Tests**: Located in `fastvideo/tests/encoders`, `fastvideo/tests/transformers`, and `fastvideo/tests/vaes`. These verify the loading and basic functionality of model components.
-* **SSIM Tests**: Located in `fastvideo/tests/ssim`. These are regression tests that compare generated videos against reference videos using the Structural Similarity Index Measure (SSIM) to detect quality degradation.
-* **Training Tests**: Located in `fastvideo/tests/training`. These validate training loops, loss calculations, and specific training techniques like LoRA, Distillation, and VSA.
-* **Inference Tests**: Located in `fastvideo/tests/inference`. These test specialized inference pipelines and optimizations (e.g., STA, V-MoBA).
-
-For now, we will focus on **SSIM Tests**.
-
-## SSIM Tests
-
-SSIM tests are located in `fastvideo/tests/ssim`. These tests generate videos using specific models and parameters, and compare them against reference videos to ensure that changes in the codebase do not degrade generation quality or alter the output unexpectedly.
-
-!!! note
-    If you are adding an SSIM test, this serves as a safeguard. Any future code changes that break or cause errors with the specific arguments and configurations you defined will trigger a failure. Therefore, it is important to include multiple settings and arguments that cover the core features of your new pipeline to ensure robust regression testing.
-
-### Directory Structure
-
-```
-fastvideo/tests/ssim/
-├── <GPU>_reference_videos/   # Reference videos organized by GPU type (e.g., L40S_reference_videos)
-│   ├── <Model_Name>/
-│   │   ├── <Backend>/        # e.g., FLASH_ATTN, TORCH_SDPA
-│   │   │   └── <Video_File>
-├── test_causal_similarity.py
-├── test_inference_similarity.py
-├── update_reference_videos.sh
-└── ...
-```
-
-### Adding a New SSIM Test
-
-To add a new SSIM test, follow these steps:
-
-1. **Create or Update a Test File**: You can add a new test function to an existing file (like `test_inference_similarity.py`) or create a new one if testing a distinct category of models.
-
-2. **Define Model Parameters**: Define the configuration for the model you want to test. This includes model path, dimensions, inference steps, and other generation parameters. **Note:** Consider using lower `num_inference_steps` or reduced resolution (e.g., 480p instead of 720p) to keep test execution time reasonable, provided it doesn't compromise the test's ability to detect regression.
-
-   ```python
-   MY_MODEL_PARAMS = {
-       "num_gpus": 1,
-       "model_path": "organization/model-name",
-       "height": 480,
-       "width": 832,
-       "num_frames": 45,
-       "num_inference_steps": 20,
-       # ... other parameters
-   }
-   ```
-
-3. **Implement the Test Function**:
-   * Use `pytest.mark.parametrize` to run the test with different prompts, backends, and models.
-   * Set the attention backend environment variable.
-   * Initialize the `VideoGenerator`.
-   * Generate the video.
-   * Compare the generated video with the reference video using `compute_video_ssim_torchvision`.
-
-   Example structure:
-
-   ```python
-   @pytest.mark.parametrize("prompt", TEST_PROMPTS)
-   @pytest.mark.parametrize("ATTENTION_BACKEND", ["FLASH_ATTN"])
-   def test_my_model_similarity(prompt, ATTENTION_BACKEND):
-       # Setup output directories
-       # ...
-
-       # Initialize Generator
-       generator = VideoGenerator.from_pretrained(...)
-       generator.generate_video(prompt, ...)
-
-       # Compare with Reference
-       ssim_values = compute_video_ssim_torchvision(
-           reference_path, generated_path, use_ms_ssim=True
-       )
-       assert ssim_values[0] >= 0.98  # Threshold
-   ```
-
-4. **Reference Videos**:
-   * When running the test for the first time (or when updating the reference), the test will fail because the reference video is missing. The generated video will be saved in `fastvideo/tests/ssim/generated_videos`.
-   * Inspect the generated video to ensure it meets quality expectations.
-   * Move the generated video to the appropriate reference folder: `fastvideo/tests/ssim/<GPU>_reference_videos/<Model>/<Backend>/`.
-   * You can use the helper script `update_reference_videos.sh` to automate copying videos from `generated_videos` to `L40S_reference_videos`. Note: Check the script to ensure paths match your environment (it defaults to `L40S_reference_videos`).
-
-### Running Tests Locally
-
-To run the SSIM tests locally:
-
-```bash
-pytest fastvideo/tests/ssim/ -vs
-```
-
-Ensure you have the necessary GPUs available as defined in your test parameters.
-
-## Modal Workflow
-
-FastVideo uses [Modal](https://modal.com/) for running tests in a CI environment. The workflow scripts are located in `fastvideo/tests/modal/`.
-
-### `pr_test.py`
-
-The main entry point for CI tests is `fastvideo/tests/modal/pr_test.py`. This script defines Modal functions that execute the pytest suites on specific hardware (e.g., L40S, H100).
-
-### Updating Modal Configuration
-
-If you add a new test that requires:
-* **Different GPU Hardware**: You may need to change the `@app.function(gpu=...)` decorator.
-* **Longer Execution Time**: Increase the `timeout` parameter.
-* **New Environment Variables/Secrets**: Add them to `secrets=[...]` or the image environment. For example, if your model is gated on Hugging Face, ensure `HF_API_KEY` is passed.
-
-For SSIM tests, the `run_ssim_tests` function in `pr_test.py` currently runs:
-
-```python
-@app.function(gpu="L40S:2", image=image, timeout=2700, secrets=[modal.Secret.from_dict({"HF_API_KEY": os.environ.get("HF_API_KEY", "")})])
-def run_ssim_tests():
-    run_test("hf auth login --token $HF_API_KEY && pytest ./fastvideo/tests/ssim -vs")
-```
-
-If your new test file is inside `fastvideo/tests/ssim`, it will automatically be picked up by this command. However, ensure that the `gpu="L40S:2"` configuration is sufficient for your model. If your model requires more GPUs (e.g., 4 or 8), you might need to create a separate Modal function or update the existing one.
-
-### Workflow Scripts
-
-The shell script that triggers these tests in the CI pipeline is located at `.buildkite/scripts/pr_test.sh`. If you add a new test category (e.g., a new folder outside of `ssim`), you will need to:
-1. Add a new function in `fastvideo/tests/modal/pr_test.py`.
-2. Add a new case in `.buildkite/scripts/pr_test.sh` to handle the new test type.
-
-!!! note
-    If you are a maintainer, you'll need to finally manually update the workflow script in Buildkite. Otherwise, a maintainer will help you update.

docs/distillation/data_preprocess.md

@@ -1,42 +0,0 @@
-
-# 🧱 Data Preprocess for Distillation
-
-For distillation, we use the same data preprocessing pipeline as training. Please refer to the [Training Data Preprocess](../training/data_preprocess.md) for general preprocessing steps.
-
-## Distillation-Specific Datasets
-
-### FastVideo 480P Synthetic Wan Dataset
-
-For Wan2.1 T2V distillation, we use the **FastVideo 480P Synthetic Wan dataset** ([FastVideo/Wan-Syn_77x448x832_600k](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x448x832_600k)) which contains 600k synthetic latents.
-
-```bash
-# Download the preprocessed dataset
-python scripts/huggingface/download_hf.py \
-    --repo_id "FastVideo/Wan-Syn_77x448x832_600k" \
-    --local_dir "FastVideo/Wan-Syn_77x448x832_600k" \
-    --repo_type "dataset"
-```
-
-### Crush Smol Dataset
-
-For Wan2.2 TI2V distillation, we use the crush_smol dataset which includes both raw videos and preprocessed latents.
-
-```bash
-# Download dataset
-python scripts/huggingface/download_hf.py \
-    --repo_id=FastVideo/mini_i2v_dataset \
-    --local_dir=data/mini_i2v_dataset \
-    --repo_type=dataset
-```
-
-## Preprocessing for Distillation
-
-The preprocessing steps are identical to training. Run the appropriate preprocessing script based on your model:
-
-```bash
-# For Wan2.1 T2V
-bash scripts/preprocess/v1_preprocess_wan_data_t2v
-
-# For Wan2.2 TI2V  
-bash examples/distill/Wan2.2-TI2V-5B-Diffusers/crush_smol/preprocess_wan_data_ti2v_5b.sh
-```

docs/distillation/dmd.md

@@ -1,87 +0,0 @@
-# 🎯 Distillation
-
-We introduce a new finetuning strategy - **Sparse-distill**, which jointly integrates **[DMD](https://arxiv.org/abs/2405.14867)** and **[VSA](https://arxiv.org/abs/2505.13389)** in a single training process. This approach combines the benefits of both distillation to shorten diffusion steps and sparse attention to reduce attention computation, enabling much faster video generation.
-
-## 📊 Model Overview
-
-We provide two distilled models:
-
-- **[FastWan2.1-T2V-1.3B-Diffusers](https://huggingface.co/FastVideo/FastWan2.1-T2V-1.3B-Diffusers)**: 3-step inference, up to **16 FPS** on H100 GPU
-- **[FastWan2.1-T2V-14B-480P-Diffusers](https://huggingface.co/FastVideo/FastWan2.1-T2V-14B-480P-Diffusers)**: 3-step inference, up to **60x speed up** at 480P, **90x speed up** at 720P for denoising loop
-- **[FastWan2.2-TI2V-5B-FullAttn-Diffusers](https://huggingface.co/FastVideo/FastWan2.2-TI2V-5B-FullAttn-Diffusers)**: 3-step inference, up to **50x speed up** at 720P for denoising loop
-
-Both models are trained on **61×448×832** resolution but support generating videos with **any resolution** (1.3B  model mainly support 480P, 14B model support 480P and 720P, quality may degrade for different resolutions).
-
-## ⚙️ Inference
-First install [VSA](../attention/vsa/index.md). Set `MODEL_BASE` to your own model path and run:
-
-```bash
-bash scripts/inference/v1_inference_wan_dmd.sh
-```
-
-## 🗂️ Dataset
-
-We use the **FastVideo 480P Synthetic Wan dataset** ([FastVideo/Wan-Syn_77x448x832_600k](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x448x832_600k)) for distillation, which contains 600k synthetic latents.
-
-### Download Dataset
-
-```bash
-# Download the preprocessed dataset
-python scripts/huggingface/download_hf.py \
-    --repo_id "FastVideo/Wan-Syn_77x448x832_600k" \
-    --local_dir "FastVideo/Wan-Syn_77x448x832_600k" \
-    --repo_type "dataset"
-```
-
-## 🚀 Training Scripts
-
-### Wan2.1 1.3B Model Sparse-Distill
-
-For the 1.3B model, we use **4 nodes with 32 H200 GPUs** (8 GPUs per node):
-
-```bash
-# Multi-node training (8 nodes, 64 GPUs total)
-sbatch examples/distill/Wan2.1-T2V/Wan-Syn-Data-480P/distill_dmd_VSA_t2v_1.3B.slurm
-```
-
-**Key Configuration:**
-- Global batch size: 64
-- Gradient accumulation steps: 2
-- Learning rate: 1e-5
-- VSA attention sparsity: 0.8
-- Training steps: 4000 (~12 hours)
-
-### Wan2.1 14B Model Sparse-Distill
-
-For the 14B model, we use **8 nodes with 64 H200 GPUs** (8 GPUs per node):
-
-```bash
-# Multi-node training (8 nodes, 64 GPUs total)
-sbatch examples/distill/Wan2.1-T2V/Wan-Syn-Data-480P/distill_dmd_VSA_t2v_14B.slurm
-```
-
-**Key Configuration:**
-- Global batch size: 64
-- Sequence parallel size: 4
-- Gradient accumulation steps: 4
-- Learning rate: 1e-5
-- VSA attention sparsity: 0.9
-- Training steps: 3000 (~52 hours)
-- HSDP shard dim: 8
-
-### Wan2.2 5B Model Sparse-Distill
-
-For the 5B model, we use **8 nodes with 64 H200 GPUs** (8 GPUs per node):
-
-```bash
-# Multi-node training (8 nodes, 64 GPUs total)
-sbatch examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free/distill_dmd_t2v_5B.sh 
-```
-
-**Key Configuration:**
-- Global batch size: 64
-- Sequence parallel size: 1
-- Gradient accumulation steps: 1
-- Learning rate: 2e-5
-- Training steps: 3000 (~12 hours)
-- HSDP shard dim: 1

docs/examples/examples_index.md

@@ -1,12 +0,0 @@
-# 💡 Examples
-
-A collection of examples demonstrating usage of FastVideo.
-
-All documented examples are autogenerated using [generate_examples.py](https://github.com/hao-ai-lab/FastVideo/blob/main/docs/generate_examples.py) from examples found in the [examples](https://github.com/hao-ai-lab/FastVideo/tree/main/examples) directory.
-
-## Examples
-
-- [Examples Distillation Index](distillation/examples/examples_distillation_index.md)
-- [Examples Training Index](training/examples/examples_training_index.md)
-- [Examples Inference Index](inference/examples/examples_inference_index.md)
-

docs/generate_examples.py

@@ -1,553 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/docs/source/generate_examples.py
-
-import itertools
-import re
-from dataclasses import dataclass, field
-from pathlib import Path
-
-ROOT_DIR = Path(__file__).parent.parent.resolve()
-ROOT_DIR_RELATIVE = '../..'
-EXAMPLE_DIR = ROOT_DIR / "examples"
-EXAMPLE_DOC_DIR = ROOT_DIR / "docs/getting_started/examples"
-GITHUB_REPO = "hao-ai-lab/FastVideo"  # Update this to your repo
-
-
-def fix_case(text: str) -> str:
-    subs = {
-        "api": "API",
-        "cli": "CLI",
-        "cpu": "CPU",
-        "llm": "LLM",
-        "tpu": "TPU",
-        "aqlm": "AQLM",
-        "gguf": "GGUF",
-        "lora": "LoRA",
-        "rlhf": "RLHF",
-        "vllm": "vLLM",
-        "openai": "OpenAI",
-        "multilora": "MultiLoRA",
-        "mlpspeculator": "MLPSpeculator",
-        "finetune": "Finetune",
-        "distillation": "Distillation",
-        "wan": "Wan",
-        "i2v": "I2V",
-        "t2v": "T2V",
-        "1.3b": "1.3B",
-        "14b": "14B",
-        "480p": "480P",
-        "720p": "720P",
-        r"fp\d+": lambda x: x.group(0).upper(),  # e.g. fp16, fp32
-        r"int\d+": lambda x: x.group(0).upper(),  # e.g. int8, int16
-    }
-    for pattern, repl in subs.items():
-        text = re.sub(rf'\b{pattern}\b', repl, text,
-                      flags=re.IGNORECASE)  # type: ignore[call-overload]
-    return text
-
-
-@dataclass
-class Index:
-    """
-    Index class to generate a structured document index.
-
-    Attributes:
-        path (Path): The path save the index file to.
-        title (str): The title of the index.
-        description (str): A brief description of the index.
-        caption (str): An optional caption for the table of contents.
-        maxdepth (int): The maximum depth of the table of contents. Defaults to 1.
-        documents (list[str]): A list of document paths to include in the index. Defaults to an empty list.
-
-    Methods:
-        generate() -> str:
-            Generates the index content as a string in the specified format.
-    """ # noqa: E501
-    path: Path
-    title: str
-    description: str
-    caption: str
-    maxdepth: int = 1
-    documents: list[str] = field(default_factory=list)
-
-    def generate(self) -> str:
-        content = f"# {self.title}\n\n{self.description}\n\n"
-        if self.caption:
-            content += f"## {self.caption}\n\n"
-        # Generate a simple list of links for MkDocs
-        for doc in self.documents:
-            # Convert document path to proper link
-            doc_link = doc.replace("\\", "/")
-            # Get just the filename for the link text
-            doc_title = fix_case(Path(doc).stem.replace("_", " ").title())
-            content += f"- [{doc_title}]({doc_link}.md)\n"
-        content += "\n"
-        return content
-
-
-@dataclass
-class Example:
-    """
-    Example class for generating documentation content from a given path.
-
-    Attributes:
-        path (Path): The path to the main directory or file.
-        category (str): The category of the document.
-        main_file (Path): The main file in the directory.
-        other_files (list[Path]): list of other files in the directory.
-        title (str): The title of the document.
-
-    Methods:
-        __post_init__(): Initializes the main_file, other_files, and title attributes.
-        determine_main_file() -> Path: Determines the main file in the given path.
-        determine_other_files() -> list[Path]: Determines other files in the directory excluding the main file.
-        determine_title() -> str: Determines the title of the document.
-        generate() -> str: Generates the documentation content.
-    """ # noqa: E501
-    path: Path
-    category: str | None = None
-    main_file: Path = field(init=False)
-    other_files: list[Path] = field(init=False)
-    title: str = field(init=False)
-
-    def __post_init__(self):
-        self.main_file = self.determine_main_file()
-        self.other_files = self.determine_other_files()
-        self.title = self.determine_title()
-
-    def determine_main_file(self) -> Path:
-        """
-        Determines the main file in the given path.
-        If the path is a file, it returns the path itself. Otherwise, it searches
-        for Markdown files (*.md) in the directory and returns the first one found.
-        Returns:
-            Path: The main file path, either the original path if it's a file or the first
-            Markdown file found in the directory.
-        Raises:
-            IndexError: If no Markdown files are found in the directory.
-        """ # noqa: E501
-        return self.path if self.path.is_file() else list(
-            self.path.glob("*.md")).pop()
-
-    def determine_other_files(self) -> list[Path]:
-        """
-        Determine other files in the directory excluding the main file.
-
-        This method checks if the given path is a file. If it is, it returns an empty list.
-        Otherwise, it recursively searches through the directory and returns a list of all
-        files that are not the main file.
-
-        Returns:
-            list[Path]: A list of Path objects representing the other files in the directory.
-        """ # noqa: E501
-        if self.path.is_file():
-            return []
-        is_other_file = lambda file: file.is_file() and file != self.main_file
-        return [file for file in self.path.rglob("*")
-                if is_other_file(file)]  # type: ignore[no-untyped-call]
-
-    def determine_title(self) -> str:
-        return fix_case(self.path.stem.replace("_", " ").title())
-
-    def generate(self) -> str:
-        # Create GitHub link to source
-        github_path = str(self.path.relative_to(ROOT_DIR)).replace("\\", "/")
-        github_url = f"https://github.com/{GITHUB_REPO}/blob/main/{github_path}"
-        content = f"**Source:** [{github_path}]({github_url})\n\n"
-
-        # Add title for code files
-        if self.main_file.suffix != ".md":
-            content += f"# {self.title}\n\n"
-
-        # Include main file content
-        if self.main_file.suffix == ".md":
-            # For markdown files, include the content directly
-            with open(self.main_file, encoding='utf-8') as f:
-                content += f.read() + "\n\n"
-        else:
-            # For code files, use code blocks
-            language = self.main_file.suffix[1:] if self.main_file.suffix else ""
-            with open(self.main_file, encoding='utf-8') as f:
-                file_content = f.read()
-            content += f"```{language}\n{file_content}\n```\n\n"
-
-        if not self.other_files:
-            return content
-
-        content += "## Additional Files\n\n"
-        # Define binary/non-text file extensions to skip
-        binary_extensions = {
-            '.mp4', '.avi', '.mov', '.mkv', '.gif', '.jpg', '.jpeg', '.png',
-            '.webp', '.bmp', '.pdf', '.zip', '.tar', '.gz', '.mp3', '.wav'
-        }
-
-        for file in sorted(self.other_files):
-            # Skip binary files
-            if file.suffix.lower() in binary_extensions:
-                continue
-
-            file_rel_path = file.relative_to(self.path)
-            # Use collapsible admonition syntax for MkDocs
-            content += f"??? note \"{file_rel_path}\"\n\n"
-
-            try:
-                if file.suffix == ".md":
-                    # Include markdown content with indentation
-                    with open(file, encoding='utf-8') as f:
-                        for line in f:
-                            content += f"    {line}"
-                else:
-                    # Include code with proper formatting
-                    language = file.suffix[1:] if file.suffix else ""
-                    with open(file, encoding='utf-8') as f:
-                        file_content = f.read()
-                    # Indent the code block for the admonition
-                    content += f"    ```{language}\n"
-                    for line in file_content.split('\n'):
-                        content += f"    {line}\n"
-                    content += "    ```\n"
-                content += "\n"
-            except UnicodeDecodeError:
-                # Skip files that can't be decoded as UTF-8
-                continue
-
-        return content
-
-
-@dataclass
-class NestedStructure:
-    """Helper class to manage nested documentation structures for training/distillation."""
-    category: str
-    method: str
-    model: str
-    dataset: str
-    example: Example
-
-    @property
-    def filename(self) -> str:
-        return f"{self.model}_{self.dataset}"
-
-    @property
-    def title(self) -> str:
-        return fix_case(self.dataset.replace('_', ' '))
-
-    @property
-    def description(self) -> str:
-        category_name = self.category.title()
-        return f"{category_name} example using the {self.dataset} dataset with the {self.model} model."
-
-
-def create_category_indices() -> dict[str, Index]:
-    """Create category indices with their respective configurations."""
-    main_index_dir = ROOT_DIR / "docs/examples"
-    if not main_index_dir.exists():
-        main_index_dir.mkdir(parents=True)
-
-    category_indices = {
-        "inference":
-        Index(
-            path=ROOT_DIR /
-            "docs/inference/examples/examples_inference_index.md",
-            title="🚀 Examples",
-            description=
-            "Inference examples demonstrate how to use FastVideo inference. We recommend starting with [basic.md](basic.md).",
-            caption="Examples",
-            maxdepth=1,
-        ),
-        "training":
-        Index(
-            path=ROOT_DIR / "docs/training/examples/examples_training_index.md",
-            title="🚀 Examples",
-            description=
-            "Training examples demonstrate how to use FastVideo training.",
-            caption="Examples",
-            maxdepth=3,
-        ),
-        "distillation":
-        Index(
-            path=ROOT_DIR /
-            "docs/distillation/examples/examples_distillation_index.md",
-            title="🚀 Examples",
-            description=
-            "Distillation examples demonstrate how to use FastVideo distillation.",
-            caption="Examples",
-            maxdepth=3,
-        ),
-    }
-
-    # Ensure all category doc directories exist
-    for index in category_indices.values():
-        if not index.path.parent.exists():
-            index.path.parent.mkdir(parents=True)
-
-    return category_indices
-
-
-def find_examples(category_indices: dict[str, Index],
-                  generate_main_index: bool) -> list[Example]:
-    """Find all examples from the examples directory."""
-    examples = []
-    glob_patterns = ["*.py", "*.md", "*.sh"]
-
-    # Map category names to actual directory names
-    category_dir_mapping = {
-        "distillation": "distill",  # examples/distill/ -> distillation category
-    }
-
-    # Find categorised examples
-    for category in category_indices:
-        # Use mapped directory name if available, otherwise use category name
-        dir_name = category_dir_mapping.get(category, category)
-        category_dir = EXAMPLE_DIR / dir_name
-
-        # Skip if directory doesn't exist
-        if not category_dir.exists():
-            continue
-
-        globs = [category_dir.glob(pattern) for pattern in glob_patterns]
-        for path in itertools.chain(*globs):
-            examples.append(Example(path, category))
-        # Find examples in subdirectories (recursively)
-        for path in category_dir.glob("**/*.md"):
-            examples.append(Example(path.parent, category))
-
-    # Find uncategorised examples only if we're generating a main index
-    if generate_main_index:
-        globs = [EXAMPLE_DIR.glob(pattern) for pattern in glob_patterns]
-        for path in itertools.chain(*globs):
-            examples.append(Example(path))
-        # Find examples in subdirectories
-        for path in EXAMPLE_DIR.glob("*/*.md"):
-            # Skip categorised examples
-            if path.parent.name in category_indices:
-                continue
-            examples.append(Example(path.parent))
-
-    return examples
-
-
-def create_nested_structures(
-    examples: list[Example]
-) -> dict[str, dict[str, dict[str, dict[str, NestedStructure]]]]:
-    """Create nested structures for training and distillation categories."""
-    nested_structures: dict[str, dict[str, dict[str,
-                                                dict[str,
-                                                     NestedStructure]]]] = {}
-
-    # Map category names to actual directory names
-    category_dir_mapping = {
-        "distillation": "distill",
-    }
-
-    for example in examples:
-        if example.category not in ["training", "distillation"]:
-            continue
-
-        # Use mapped directory name if available
-        dir_name = category_dir_mapping.get(example.category, example.category)
-        category_dir = EXAMPLE_DIR / dir_name
-        relative_path = example.path.relative_to(category_dir)
-        path_parts = relative_path.parts
-
-        if example.category == "training":
-            # For training examples like finetune/wan_i2v_14b_480p/crush_smol
-            if len(path_parts) >= 3:
-                method = path_parts[0]  # e.g., "finetune"
-                model = path_parts[1]  # e.g., "wan_i2v_14b_480p"
-                dataset = path_parts[2]  # e.g., "crush_smol"
-
-                # Initialize nested structure
-                if example.category not in nested_structures:
-                    nested_structures[example.category] = {}
-                if method not in nested_structures[example.category]:
-                    nested_structures[example.category][method] = {}
-                if model not in nested_structures[example.category][method]:
-                    nested_structures[example.category][method][model] = {}
-
-                # Store the nested structure
-                nested_structures[
-                    example.category][method][model][dataset] = NestedStructure(
-                        category=example.category,
-                        method=method,
-                        model=model,
-                        dataset=dataset,
-                        example=example)
-
-        elif example.category == "distillation" and len(path_parts) >= 2:
-            # For distillation examples like Wan2.1-T2V/Wan-Syn-Data-480P
-            model = path_parts[0]  # e.g., "Wan2.1-T2V"
-            dataset = path_parts[1]  # e.g., "Wan-Syn-Data-480P"
-            method = "DMD"  # Default method for distillation
-
-            # Initialize nested structure
-            if example.category not in nested_structures:
-                nested_structures[example.category] = {}
-            if method not in nested_structures[example.category]:
-                nested_structures[example.category][method] = {}
-            if model not in nested_structures[example.category][method]:
-                nested_structures[example.category][method][model] = {}
-
-            # Store the nested structure
-            nested_structures[
-                example.category][method][model][dataset] = NestedStructure(
-                    category=example.category,
-                    method=method,
-                    model=model,
-                    dataset=dataset,
-                    example=example)
-
-    return nested_structures
-
-
-def generate_flat_examples(examples: list[Example],
-                           category_indices: dict[str, Index],
-                           examples_index: Index | None,
-                           generate_main_index: bool) -> None:
-    """Generate documentation for flat structure examples (inference, etc.)."""
-    for example in examples:
-        if example.category in ["training", "distillation"]:
-            continue  # Skip nested structure examples
-
-        # Determine which index to use for this example
-        if example.category is not None and example.category in category_indices:
-            index = category_indices[example.category]
-        elif generate_main_index:
-            assert examples_index is not None
-            index = examples_index
-        else:
-            continue
-
-        # Generate the example documentation
-        doc_path = index.path.parent / f"{example.path.stem}.md"
-        with open(doc_path, "w+") as f:
-            f.write(example.generate())
-        index.documents.append(example.path.stem)
-
-
-def generate_nested_examples(nested_structures: dict[str, dict[str, dict[
-    str, dict[str, NestedStructure]]]], category_indices: dict[str,
-                                                               Index]) -> None:
-    """Generate documentation for nested structure examples (training, distillation)."""
-    for category_name in ["training", "distillation"]:
-        if category_name not in category_indices or category_name not in nested_structures:
-            continue
-
-        category_index = category_indices[category_name]
-        category_base_dir = category_index.path.parent
-
-        for method, models in nested_structures[category_name].items():
-            # Create method-level index
-            method_index = Index(path=category_base_dir / f"{method}.md",
-                                 title=fix_case(method),
-                                 description=f"Examples using {method}.",
-                                 caption=f"{fix_case(method)} Examples",
-                                 maxdepth=2)
-
-            for model, datasets in models.items():
-                # Generate dataset examples using the Example class
-                for dataset, nested_struct in datasets.items():
-                    doc_path = category_base_dir / f"{nested_struct.filename}.md"
-                    with open(doc_path, "w+") as f:
-                        f.write(nested_struct.example.generate())
-
-                # Create model-level index
-                model_index = Index(
-                    path=category_base_dir / f"{model}.md",
-                    title=fix_case(model.replace('_', ' ')),
-                    description=f"Examples for the {model} model.",
-                    caption=f"{fix_case(model.replace('_', ' '))} Datasets",
-                    maxdepth=1)
-
-                # Add dataset indices to model index
-                for dataset, nested_struct in datasets.items():
-                    model_index.documents.append(nested_struct.filename)
-
-                # Write model index
-                with open(model_index.path, "w+") as f:
-                    f.write(model_index.generate())
-
-                # Add model to method index
-                method_index.documents.append(model)
-
-            # Write method index
-            with open(method_index.path, "w+") as f:
-                f.write(method_index.generate())
-
-            # Add method to main category index
-            category_index.documents.append(method)
-
-
-def generate_examples(generate_main_index: bool = False) -> None:
-    """
-    Generate example documentation.
-    
-    Args:
-        generate_main_index (bool): Whether to generate the main examples index.
-            If False, only category-specific indices will be generated.
-    """
-    # Create category indices
-    category_indices = create_category_indices()
-
-    # Create the main examples index only if requested
-    examples_index = None
-    if generate_main_index:
-        main_index_dir = ROOT_DIR / "docs/examples"
-        examples_index = Index(
-            path=main_index_dir / "examples_index.md",
-            title="💡 Examples",
-            description=
-            "A collection of examples demonstrating usage of FastVideo.\n\n"
-            f"All documented examples are autogenerated using [generate_examples.py](https://github.com/{GITHUB_REPO}/blob/main/docs/generate_examples.py) "
-            f"from examples found in the [examples](https://github.com/{GITHUB_REPO}/tree/main/examples) directory.",
-            caption="Examples",
-            maxdepth=2)
-
-    # Find all examples
-    examples = find_examples(category_indices, generate_main_index)
-
-    # Create nested structures for training and distillation
-    nested_structures = create_nested_structures(examples)
-
-    # Generate flat structure examples (inference, etc.)
-    generate_flat_examples(examples, category_indices, examples_index,
-                           generate_main_index)
-
-    # Generate nested structure examples (training, distillation)
-    generate_nested_examples(nested_structures, category_indices)
-
-    # Generate the index files for categories
-    for category_index in category_indices.values():
-        if category_index.documents:
-            # Add to main index if it exists
-            if generate_main_index and examples_index:
-                main_index_dir = examples_index.path.parent
-                rel_path = category_index.path.relative_to(
-                    main_index_dir.parent)
-                examples_index.documents.insert(
-                    0,
-                    str(rel_path).replace(".md", ""))
-
-            # Write the category index file
-            with open(category_index.path, "w+") as f:
-                f.write(category_index.generate())
-
-    # Write the main index file if requested
-    if generate_main_index and examples_index:
-        with open(examples_index.path, "w+") as f:
-            f.write(examples_index.generate())
-
-
-def on_pre_build_hook(config, **kwargs):
-    """
-    MkDocs hook to generate examples before building the documentation.
-    This function is called automatically by the mkdocs-simple-hooks plugin.
-    """
-    print("Generating example documentation...")
-    generate_examples(generate_main_index=True)
-    print("Example documentation generated successfully!")
-
-
-if __name__ == "__main__":
-    print("Generating example documentation...")
-    generate_examples(generate_main_index=True)
-    print("Example documentation generated successfully!")

docs/getting_started/installation.md

@@ -1,41 +0,0 @@
-
-# 🔧 Installation
-
-FastVideo supports the following hardware platforms:
-
-- [NVIDIA CUDA](installation/gpu.md)
-- [Apple silicon](installation/mps.md)
-
-## Quick Installation
-
-### Using pip
-
-```bash
-pip install fastvideo
-```
-
-### Using conda
-
-```bash
-conda install -c conda-forge fastvideo
-```
-
-### From source
-
-```bash
-git clone https://github.com/hao-ai-lab/FastVideo.git
-cd FastVideo
-pip install -e .
-```
-
-## Hardware Requirements
-
-- **NVIDIA GPUs**: CUDA 11.8+ with compute capability 7.0+
-- **Apple Silicon**: macOS 12.0+ with M1/M2/M3 chips
-- **CPU**: x86_64 architecture (for CPU-only inference)
-
-## Next Steps
-
-- [Quick Start Guide](quick_start.md) - Get started with your first video generation
-- [Configuration](../inference/configuration.md) - Learn about configuration options
-- [Examples](../inference/examples/examples_inference_index.md) - Explore example scripts and notebooks

docs/getting_started/installation/mps.md

@@ -1,93 +0,0 @@
-# MPS (Apple Silicon)
-
-Instructions to install FastVideo for Apple Silicon.
-
-## Requirements
-
-- **OS: MacOS**
-- **Python: 3.12.4**
-
-## Set up using Python
-
-### Create a new Python environment
-
-#### Conda
-
-You can create a new python environment using [Conda](https://docs.conda.io/projects/conda/en/stable/user-guide/getting-started.html)
-
-##### 1. Install Miniconda (if not already installed)
-
-```bash
-wget https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh
-bash Miniconda3-latest-MacOSX-arm64.sh
-source ~/.zshrc
-```
-
-##### 2. Create and activate a Conda environment for FastVideo
-
-```bash
-# (Recommended) Create a new conda environment.
-conda create -n fastvideo python=3.12.4 -y
-conda activate fastvideo
-```
-
-#### uv
-
-Or you can create a new Python environment using [uv](https://docs.astral.sh/uv/), a very fast Python environment manager. 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 command:
-
-```console
-# (Recommended) Create a new uv environment. Use `--seed` to install `pip` and `setuptools` in the environment.
-uv venv --python 3.12 --seed
-source .venv/bin/activate
-```
-
-### Dependencies
-
-```
-brew install ffmpeg
-```
-
-### Installation
-
-```bash
-pip install fastvideo
-
-# or if you are using uv
-uv pip install fastvideo
-```
-
-### Installation from Source
-
-#### 1. Clone the FastVideo repository
-
-```bash
-git clone https://github.com/hao-ai-lab/FastVideo.git && cd FastVideo
-```
-
-#### 2. Install FastVideo
-
-Basic installation:
-
-```bash
-pip install -e .
-
-# or if you are using uv
-uv pip install -e .
-```
-
-## Development Environment Setup
-
-If you're planning to contribute to FastVideo please see the following page:
-[Contributor Guide](../../contributing/overview.md)
-
-## Hardware Requirements
-
-### For Basic Inference
-
-- Mac M1, M2, M3, or M4 (at least 32 GB RAM is preferable for high quality video generation)
-
-## Troubleshooting
-
-If you encounter any issues during installation, please open an issue on our [GitHub repository](https://github.com/hao-ai-lab/FastVideo).
-
-You can also join our [Slack community](https://join.slack.com/t/fastvideo/shared_invite/zt-38u6p1jqe-yDI1QJOCEnbtkLoaI5bjZQ) for additional support.

docs/getting_started/quick_start.md

@@ -1,77 +0,0 @@
-# 🚀 Quick Start
-
-Get up and running with FastVideo in minutes!
-
-## Installation
-
-First, install FastVideo:
-
-```bash
-# Create and activate a new conda environment
-conda create -n fastvideo python=3.12
-conda activate fastvideo
-
-# Install FastVideo
-pip install fastvideo
-```
-
-## Basic Usage
-
-### Text-to-Video Generation
-
-```python
-from fastvideo import VideoGenerator
-
-def main():
-    # Create a video generator with a pre-trained model
-    generator = VideoGenerator.from_pretrained(
-        "Wan-AI/Wan2.1-T2V-1.3B-Diffusers",
-        num_gpus=1,  # Adjust based on your hardware
-    )
-
-    # Define a prompt for your video
-    prompt = "A curious raccoon peers through a vibrant field of yellow sunflowers, its eyes wide with interest."
-
-    # Generate the video
-    video = generator.generate_video(
-        prompt,
-        return_frames=True,  # Also return frames from this call (defaults to False)
-        output_path="my_videos/",  # Controls where videos are saved
-        save_video=True
-    )
-
-if __name__ == '__main__':
-    main()
-```
-
-### Image-to-Video Generation
-
-```python
-from fastvideo import VideoGenerator, SamplingParam
-
-def main():
-    # Create the generator
-    model_name = "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers"
-    generator = VideoGenerator.from_pretrained(model_name, num_gpus=1)
-
-    # Set up parameters with an initial image
-    sampling_param = SamplingParam.from_pretrained(model_name)
-    sampling_param.image_path = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/astronaut.jpg"
-    sampling_param.num_frames = 107
-
-    # Generate video based on the image
-    prompt = "A photograph coming to life with gentle movement"
-    generator.generate_video(prompt, sampling_param=sampling_param,
-                             output_path="my_videos/",
-                             save_video=True)
-
-if __name__ == '__main__':
-    main()
-```
-
-## Next Steps
-
-- [Installation Guide](installation.md) - Detailed installation instructions
-- [Configuration](../inference/configuration.md) - Learn about configuration options
-- [Examples](../inference/examples/) - Explore more examples
-- [Optimizations](../inference/optimizations.md) - Performance optimization tips

docs/getting_started/v1_api.md

@@ -1,42 +0,0 @@
-# V1 API
-
-FastVideo's V1 API provides a streamlined interface for video generation tasks with powerful customization options. This page documents the primary components of the API.
-
-## Video Generator
-
-This class will be the primary Python API for generating videos and images.
-
-::: fastvideo.entrypoints.video_generator.VideoGenerator
-    options:
-      show_root_heading: true
-      show_source: false
-      members:
-        - from_pretrained
-      heading_level: 3
-
-`VideoGenerator.from_pretrained()` should be the primary way of creating a new video generator.
-
-## Configuring FastVideo
-
-The following two classes `PipelineConfig` and `SamplingParam` are used to configure initialization and sampling parameters, respectively.
-
-### PipelineConfig
-
-::: fastvideo.configs.pipelines.base.PipelineConfig
-    options:
-      show_root_heading: true
-      show_source: false
-      members:
-        - from_pretrained
-        - dump_to_json
-      heading_level: 4
-
-### SamplingParam
-
-::: fastvideo.configs.sample.base.SamplingParam
-    options:
-      show_root_heading: true
-      show_source: false
-      members:
-        - from_pretrained
-      heading_level: 4