deploy to readthedocs

.github/scripts/runpod_api.py

@@ -53,7 +53,8 @@ def create_pod():
         "volumeInGb": args.volume_size,
         "gpuTypeIds": [args.gpu_type],
         "gpuCount": args.gpu_count,
-        "imageName": args.image
+        "imageName": args.image,
+        "allowedCudaVersions": ["12.4"]
     }
 
     response = requests.post(PODS_API, headers=HEADERS, json=payload)
@@ -170,7 +171,7 @@ def execute_command(pod_id):
                                    stdout=subprocess.PIPE,
                                    stderr=subprocess.STDOUT,
                                    universal_newlines=True,
-                                   bufsize=1)
+                                   bufsize=0)
 
         stdout_lines = []
 

.github/workflows/docs.yml

@@ -0,0 +1,77 @@
+# 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
+    paths:
+      - "docs/**/*.md"
+  pull_request:
+    branches: 
+      - main
+    types: [opened, ready_for_review, synchronize, reopened]
+    paths:
+      - "docs/**/*.md"
+
+  # 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:
+  # Build job
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Setup Pages
+        id: pages
+        uses: actions/configure-pages@v5
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.10"
+      - name: Install dependencies
+        run: |
+          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: ./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
+    steps:
+      - name: Deploy to GitHub Pages
+        id: deployment
+        uses: actions/deploy-pages@v4

.readthedocs.yaml

@@ -0,0 +1,21 @@
+# Read the Docs configuration file
+# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
+
+version: 2
+
+build:
+  os: ubuntu-22.04
+  tools:
+    python: "3.12"
+
+sphinx:
+  configuration: docs/source/conf.py
+  fail_on_warning: true
+
+# If using Sphinx, optionally build your docs in additional formats such as PDF
+formats: []
+
+# Optionally declare the Python requirements required to build your docs
+python:
+  install:
+    - requirements: docs/requirements-docs.txt

docs/source/conf.py

@@ -17,6 +17,7 @@ import inspect
 import logging
 import os
 import sys
+from typing import Optional
 
 import requests
 from sphinx.ext import autodoc
@@ -136,7 +137,8 @@ _cached_base: str = ""
 _cached_branch: str = ""
 
 
-def get_repo_base_and_branch(pr_number):
+def get_repo_base_and_branch(
+        pr_number: str) -> tuple[Optional[str], Optional[str]]:
     global _cached_base, _cached_branch
     if _cached_base and _cached_branch:
         return _cached_base, _cached_branch
@@ -158,7 +160,6 @@ def linkcode_resolve(domain, info):
         return None
     if not info['module']:
         return None
-    filename = info['module'].replace('.', '/')
     module = info['module']
 
     # try to determine the correct file and line number to link to
@@ -173,7 +174,7 @@ def linkcode_resolve(domain, info):
 
             if not (inspect.isclass(obj) or inspect.isfunction(obj)
                     or inspect.ismethod(obj)):
-                obj = obj.__class__  # Get the class of the instance
+                obj = obj.__class__  # type: ignore[assignment]
 
             lineno = inspect.getsourcelines(obj)[1]
             filename = (inspect.getsourcefile(obj)

docs/source/developer_guide/overview.md

@@ -0,0 +1,50 @@
+# Contributing to FastVideo
+
+Thank you for your interest in contributing to FastVideo. We want to make the process as smooth for you as possible and this is a guide to help get you started!
+
+Our community is open to everyone and welcomes any contributions no matter how large or small.
+
+# Developer Environment:
+Do make sure you have CUDA 12.4 installed and supported. FastVideo currently only support Linux and CUDA GPUs, but we hope to support other platforms in the future.
+
+We recommend using a fresh Python 3.10 Conda environment to develop FastVideo:
+
+Install Miniconda:
+
+```
+wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
+bash Miniconda3-latest-Linux-x86_64.sh
+source ~/.bashrc
+```
+
+Create and activate a Conda environment for FastVideo:
+
+```
+conda create -n fastvideo python=3.10 -y
+conda activate fastvideo
+```
+
+Clone the FastVideo repository and go to the FastVideo directory:
+
+```
+git clone https://github.com/vllm-project/vllm.git && cd vllm
+
+```
+
+Now you can install FastVideo and setup git hooks for running linting. By using `pre-commit`, the linters will run and have to pass before you'll be able to make a commit.
+
+```bash
+pip install -e .[dev]
+
+# Can also install flash-attn (optional)
+pip install flash-attn==2.7.0.post2 --no-build-isolation 
+
+# Linting, formatting and static type checking
+pre-commit install --hook-type pre-commit --hook-type commit-msg
+
+# You can manually run pre-commit with
+pre-commit run --all-files
+
+# Unit tests
+pytest tests/
+```

docs/source/generate_examples.py

@@ -4,6 +4,7 @@ import itertools
 import re
 from dataclasses import dataclass, field
 from pathlib import Path
+from typing import Optional
 
 ROOT_DIR = Path(__file__).parent.parent.parent.resolve()
 ROOT_DIR_RELATIVE = '../../../..'
@@ -30,7 +31,8 @@ def fix_case(text: str) -> str:
         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)
+        text = re.sub(rf'\b{pattern}\b', repl, text,
+                      flags=re.IGNORECASE)  # type: ignore[call-overload]
     return text
 
 
@@ -86,7 +88,7 @@ class Example:
         generate() -> str: Generates the documentation content.
     """ # noqa: E501
     path: Path
-    category: str = None
+    category: Optional[str] = None
     main_file: Path = field(init=False)
     other_files: list[Path] = field(init=False)
     title: str = field(init=False)
@@ -124,7 +126,8 @@ class Example:
         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)]
+        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())
@@ -139,7 +142,7 @@ class Example:
             "literalinclude"
         if include == "literalinclude":
             content += f"# {self.title}\n\n"
-        content += f":::{{{include}}} {make_relative(self.main_file)}\n"
+        content += f":::{{{include}}} {make_relative(self.main_file)}\n"  # type: ignore[no-untyped-call]
         if include == "literalinclude":
             content += f":language: {self.main_file.suffix[1:]}\n"
         content += ":::\n\n"
@@ -152,7 +155,7 @@ class Example:
             include = "include" if file.suffix == ".md" else "literalinclude"
             content += f":::{{admonition}} {file.relative_to(self.path)}\n"
             content += ":class: dropdown\n\n"
-            content += f":::{{{include}}} {make_relative(file)}\n:::\n"
+            content += f":::{{{include}}} {make_relative(file)}\n:::\n"  # type: ignore[no-untyped-call]
             content += ":::\n\n"
 
         return content
@@ -228,6 +231,7 @@ def generate_examples():
         with open(doc_path, "w+") as f:
             f.write(example.generate())
         # Add the example to the appropriate index
+        assert example.category is not None
         index = category_indices.get(example.category, examples_index)
         index.documents.append(example.path.stem)
 

docs/source/getting_started/installation.md

@@ -2,7 +2,9 @@
 
 # 🔧 Installation
 The code is tested on Python 3.10.0, CUDA 12.4 and H100.
+
 ```
 ./env_setup.sh fastvideo
 ```
-To try Sliding Tile Attention (optional), please follow the instruction in [here](#sta-installation) to install STA.
+
+To try Sliding Tile Attention (optional), please follow the instruction in [here](#sta-installation) to install STA.

docs/source/index.md

@@ -75,6 +75,13 @@ inference/fasthunyuan
 inference/fastmochi
 :::
 
+:::{toctree}
+:caption: Developer Guide
+:maxdepth: 1
+
+developer_guide/overview
+:::
+
 ## Indices and tables
 
 - {ref}`genindex`

docs/source/inference/fasthunyuan.md

@@ -3,30 +3,31 @@
 # FastHunyuan
 ## Inference FastHunyuan on single RTX4090
 We now support NF4 and LLM-INT8 quantized inference using BitsAndBytes for FastHunyuan. With NF4 quantization, inference can be performed on a single RTX 4090 GPU, requiring just 20GB of VRAM.
+
 ```bash
 # Download the model weight
 python scripts/huggingface/download_hf.py --repo_id=FastVideo/FastHunyuan-diffusers --local_dir=data/FastHunyuan-diffusers --repo_type=model
 # CLI inference
 bash scripts/inference/inference_hunyuan_hf_quantization.sh
 ```
-For more information about the VRAM requirements for BitsAndBytes quantization, please refer to the table below (timing measured on an H100 GPU):
 
+For more information about the VRAM requirements for BitsAndBytes quantization, please refer to the table below (timing measured on an H100 GPU):
 
 | Configuration                  | Memory to Init Transformer | Peak Memory After Init Pipeline (Denoise) | Diffusion Time | End-to-End Time |
 |--------------------------------|----------------------------|--------------------------------------------|----------------|-----------------|
 | BF16 + Pipeline CPU Offload    | 23.883G                   | 33.744G                                    | 81s            | 121.5s          |
 | INT8 + Pipeline CPU Offload    | 13.911G                   | 27.979G                                    | 88s            | 116.7s          |
 | NF4 + Pipeline CPU Offload     | 9.453G                    | 19.26G                                     | 78s            | 114.5s          |
-           
-
-
+  
 For improved quality in generated videos, we recommend using a GPU with 80GB of memory to run the BF16 model with the original Hunyuan pipeline. To execute the inference, use the following section:
 
 ## FastHunyuan
+
 ```bash
 # Download the model weight
 python scripts/huggingface/download_hf.py --repo_id=FastVideo/FastHunyuan --local_dir=data/FastHunyuan --repo_type=model
 # CLI inference
 bash scripts/inference/inference_hunyuan.sh
 ```
-You can also inference FastHunyuan in the [official Hunyuan github](https://github.com/Tencent/HunyuanVideo).
+
+You can also inference FastHunyuan in the [official Hunyuan github](https://github.com/Tencent/HunyuanVideo).

docs/source/inference/fastmochi.md

@@ -6,4 +6,4 @@
 # Download the model weight
 python scripts/huggingface/download_hf.py --repo_id=FastVideo/FastMochi-diffusers --local_dir=data/FastMochi-diffusers --repo_type=model
 # CLI inference
-bash scripts/inference/inference_mochi_sp.sh
+bash scripts/inference/inference_mochi_sp.sh

docs/source/inference/hunyuanvideo.md

@@ -3,12 +3,16 @@
 # HunyuanVideo
 ## Inference HunyuanVideo with Sliding Tile Attention
 First, download the model:
+
 ```bash
 python scripts/huggingface/download_hf.py --repo_id=FastVideo/hunyuan --local_dir=data/hunyuan --repo_type=model 
 ```
+
 We provide two examples in the following script to run inference with STA + [TeaCache](https://github.com/ali-vilab/TeaCache) and STA only.
+
 ```bash
 sh scripts/inference/inference_hunyuan_STA.sh
 ```
+
 ## Video Demos using STA + Teacache
 Visit our [demo website](https://fast-video.github.io/) to explore our complete collection of examples. We shorten a single video generation process from 945s to 317s on H100.

docs/source/inference/stepvideo.md

@@ -1,13 +1,16 @@
 (stepvideo)=
 
 # StepVideo
-## Inference StepVideo with Sliding Tile Attention 
+## Inference StepVideo with Sliding Tile Attention
 First, download the model:
+
 ```
 python scripts/huggingface/download_hf.py --repo_id=stepfun-ai/stepvideo-t2v --local_dir=data/stepvideo-t2v --repo_type=model 
 ```
+
 Use the following scripts to run inference for StepVideo. When using STA for inference, the generated videos will have dimensions of 204×768×768 (currently, this is the only supported shape).
+
 ```bash
 sh scripts/inference/inference_stepvideo_STA.sh # Inference stepvideo with STA
 sh scripts/inference/inference_stepvideo.sh # Inference original stepvideo
-```
+```

docs/source/sliding_tile_attention/demo.md

@@ -1,11 +1,11 @@
 (sta-demo)=
 
 # Demo
-There is a demo for 2D STA with window size (6,6) operating on a (10, 10) image. 
+There is a demo for 2D STA with window size (6,6) operating on a (10, 10) image.
 
 <div style="text-align: center;">
   <video controls width="800">
     <source src="https://github.com/user-attachments/assets/f3b6dd79-7b43-4b60-a0fa-3d6495ec5747" type="video/mp4">
     Your browser does not support the video tag.
   </video>
-</div>
+</div>

docs/source/sliding_tile_attention/installation.md

@@ -13,11 +13,13 @@ sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave
 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
-```
+```

docs/source/sliding_tile_attention/test.md

@@ -1,6 +1,7 @@
 (sta-test)=
 
 # Test
+
 ```bash
 python test/test_sta.py
-```
+```

docs/source/sliding_tile_attention/usage.md

@@ -14,4 +14,4 @@ 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)
 
-```
+```

pyproject.toml

@@ -56,8 +56,8 @@ lint = [
 ]
 
 test = [ 
-    "av",
-    "pytorch-msssim",
+    "av==14.3.0",
+    "pytorch-msssim==1.0.0",
     "pytest",
 ]