hao test

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/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)
 
-```
+```