[docs] fix FastH3 installation instructions

README.md

@@ -3,7 +3,7 @@ license: other
 license_name: minimax-h3-community
 license_link: LICENSE
 base_model: MiniMaxAI/MiniMax-H3
-library_name: diffusers
+library_name: fastvideo
 pipeline_tag: text-to-video
 tags:
 - text-to-video
@@ -37,24 +37,33 @@ was trained with data-free DMD2.
 
 ## Run with FastVideo
 
+Install [uv](https://docs.astral.sh/uv/getting-started/installation/), then use
+the CUDA 13 / Blackwell path below. It selects FastVideo's published CUDA
+kernel wheel instead of compiling the kernel locally. See the
+[installation guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/)
+for other platforms.
+
 ```bash
 git clone https://github.com/hao-ai-lab/FastVideo.git
 cd FastVideo
-uv venv --python 3.12
+uv venv --python 3.12 --seed
 source .venv/bin/activate
-UV_TORCH_BACKEND=cu130 uv pip install -e ".[fasth3]"
+UV_TORCH_BACKEND=cu130 uv pip install \
+  --no-sources-package fastvideo-kernel \
+  -e ".[fasth3]"
 ```
 
 ```bash
-FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN \
+FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN FASTVIDEO_FA4=1 \
 python examples/inference/basic/basic_minimax_h3_t2v.py \
   --model-path FastVideo/FastVideo-FastH3-4-step-Preview-v1-Dense-DataFree \
   --prompt "your prompt" \
   --steps 5
 ```
 
-Five scheduler points execute the trained four transformer forwards. Adjust
-`--num-gpus` for your setup.
+Five scheduler points execute the trained four transformer forwards. The
+default uses four GPUs; other supported GPU counts must divide H3's 56
+attention heads.
 
 ## Scope