[docs] simplify FastH3 model card

README.md

@@ -25,84 +25,43 @@ tags:
 
 # FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree
 
-FastH3 4-Step Preview v1 checkpoint from FastVideo. This variant is the v10.5
-data-free DMD2 MiniMax-H3 student at training step 1300. It generates
-synchronized video and audio with four transformer forwards and uses VSA-H3 at
-90% sparsity.
+The recommended FastH3 Preview v1 checkpoint from
+[FastVideo](https://github.com/hao-ai-lab/FastVideo). It generates synchronized
+video and audio from text with four transformer forwards. This step-1300 model
+was trained with data-free DMD2 and VSA-H3 at 90% sparsity.
 
-See the complete [FastH3 4-Step Preview v1
-collection](https://huggingface.co/collections/FastVideo/fastvideo-fasth3).
+[Blog](https://haoailab.com/blogs/fasth3-preview/) ·
+[Matching LoRA](https://huggingface.co/FastVideo/FastVideo-FastH3-4-step-Preview-v1-LoRA/tree/main/vsa-datafree) ·
+[FastH3 collection](https://huggingface.co/collections/FastVideo/fastvideo-fasth3)
 
-> **VSA requirement:** This checkpoint must be run with FastVideo's Video
-> Sparse Attention (VSA-H3) backend and kernel enabled at 90% sparsity with
-> 64-token tiles. Dense attention is not a supported substitute.
+> This checkpoint requires FastVideo's VSA-H3 attention backend. Use the
+> matching LoRA above if you prefer to download only the distilled adapter.
 
-## Checkpoint identity
-
-- Training run: `20260823T132231Z-v10p5-datafree-sp4`
-- W&B: [a1etree2](https://wandb.ai/wlsaidhi/h3-dmd2-vsa/runs/a1etree2)
-- Step: 1300
-- Transformer content SHA-256: `b36987515e4c75fa4c7aaa632a7842c829ea141b235358a54d782b51230497b3`
-- Checkpoint metadata SHA-256: `dcad0fbee2a7c7e75e53435f4fd98fccf3138844883874edf057962ab48fa428`
-- FastVideo execution commit: `48a047c05ff4138f20cfa33351499c6ec5945f5d`
-
-`checkpoint_content.json` records every indexed transformer shard and digest.
-`checkpoint_metadata.json` preserves the source training configuration.
-
-## Contents
-
-This is a self-contained Diffusers modular-pipeline package. The distilled
-bf16 student is under `transformer/`; the text encoder, tokenizer, processor,
-video VAE, audio VAE, and schedulers are unmodified MiniMax-H3 components.
-
-The `transformer_ref` component is not bundled because this student targets
-T2AV. Its modular-index entry remains on `MiniMaxAI/MiniMax-H3`.
-
-## Exact FastVideo inference contract
-
-The student was trained on the explicit ladder `[999, 749, 500, 250]`. These
-are four transformer forwards; do not replace the ladder with an arbitrary
-four-point scheduler grid.
+## Run with FastVideo
 
 ```bash
-export FASTVIDEO_DMD_DENOISING_STEPS=999,749,500,250
+git clone https://github.com/hao-ai-lab/FastVideo.git
+cd FastVideo
+uv venv --python 3.12
+source .venv/bin/activate
+UV_TORCH_BACKEND=cu130 uv pip install -e ".[fasth3]"
+```
 
+```bash
 python examples/inference/basic/basic_fasth3.py \
   --model-path FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree \
-  --prompt '<MiniMax-H3 T2AV prompt>' \
-  --steps 5 \
-  --num-gpus 4 \
-  --vsa-sparsity 0.9 \
-  --vsa-tile-size 64 \
-  --vsa-kernel sm100a
+  --prompt "your prompt" \
+  --no-warmup \
+  --repeats 1
 ```
 
-The five scheduler grid points execute the four trained jumps. Use guidance
-scale 1.0, VSA-H3 at 90% sparsity with 64-token tiles, and SP=4/FSDP across one
-four-GPU B200 tray for the validated route. The exact v10.5 execution commit
-runs the sparse DiT eagerly; regional compile is disabled for this snapshot.
-
-## Training summary
-
-- Method: data-free DMD2 with carried backward-simulation rollouts.
-- Student grid: `[999, 749, 500, 250]`.
-- Student attention: VSA-H3, 64-token tiles, 90% sparsity.
-- Prompt conditioning: mixed native resolution and duration T2AV prompts; no
-  target video latents are consumed by the data-free objective.
-- Global batch size: 64.
-- Student and critic learning rate: `2e-6`.
-- Precision: fp32 training state and bf16 inference export.
-- Training topology: 32 B200 GPUs, SP=4, full-world FSDP sharding.
+The script uses FastVideo's optimized four-GPU defaults and the trained
+four-forward schedule. On systems without the Blackwell kernels, add
+`--vsa-kernel triton --no-fa4`; adjust `--num-gpus` for your setup.
 
-## Limitations
+## Scope
 
-- Intermediate checkpoint at step 1300 of a planned 4000-step run.
-- This historical checkpoint predates the later continuous score-clock
-  and FastGen-alignment corrections. Use it to reproduce v10.5 results, not as
-  the reference implementation of the corrected training recipe.
-- T2AV only; the reference-conditioned transformer was not distilled here.
-- Quality remains below base MiniMax-H3 on difficult motion, fine structure,
-  and some audio details.
-- Sampling outside the trained ladder or attention geometry is
-  off-distribution.
-- Inherits the MiniMax H3 Community License and base-model limitations.
+This preview supports text-to-audio-video generation. FL2VA and Ref2VA were
+not distilled. Difficult motion, fine detail, and some audio may remain below
+the base MiniMax H3 model. This checkpoint inherits the
+[MiniMax H3 Community License](LICENSE).