Add INT8 MLX model card and metadata

149946b verified 2 days ago

2.86 kB

license: other
license_name: minimax-h3-community
license_link: >-
  https://huggingface.co/FastVideo/FastVideo-FastH3-4-step-Preview-v1-Dense-DataFree/blob/main/LICENSE
base_model: FastVideo/FastVideo-FastH3-4-step-Preview-v1-Dense-DataFree
library_name: fastvideo
pipeline_tag: text-to-video
tags:
  - mlx
  - apple-silicon
  - minimax-h3
  - fasth3
  - text-to-video
  - audio
  - quantized
  - int8

FastH3 Preview v1 Dense DataFree MLX INT8

This repository contains the ready-to-run INT8 MLX DiT for FastVideo/FastVideo-FastH3-4-step-Preview-v1-Dense-DataFree. It avoids the local conversion step and its temporary unified-memory requirement on Apple silicon.

The checkpoint is the dense-attention, step-1000, four-forward Preview v1 model. Quantization is affine, weight-only INT8 with group size 64. Activations remain BF16. This export is dense-only and does not support --vsa.

Files

File	Purpose
`mlx_h3_dit.safetensors`	Quantized H3 DiT weights, 20.07 GiB
`mlx_h3_dit.json`	Architecture, quantization, and cached AdaLN schedule
`conversion_manifest.json`	Source revision, converter version, validation, and checksums

Weights SHA-256: 9821dd3386f13b5a4dc11ba12f37bf51aada227a319471c409066e79f7d2f59c

Download

Download the shared H3 components once. They include the tokenizer, Qwen3-VL text encoder, video VAE, and audio VAE:

hf download FastVideo/FastVideo-FastH3-4-step-Preview-v1-Dense-DataFree \
  --revision f624f08c6c279ab43534c003e556fc5b295b6558 \
  --local-dir ./FastH3-Preview-v1-Dense-DataFree

Then download this ready-to-run MLX DiT:

hf download FastVideo/FastVideo-FastH3-4-step-Preview-v1-Dense-DataFree-MLX-INT8 \
  --local-dir ./FastH3-MLX-INT8

Run on Apple silicon

From a current FastVideo checkout:

uv venv --python 3.12 --seed
source .venv/bin/activate
uv pip install -e ".[mlx]"

python examples/inference/basic/mlx_fasth3.py \
  --model-root ./FastH3-Preview-v1-Dense-DataFree \
  --mlx-checkpoint ./FastH3-MLX-INT8 \
  --prompt '(S1) A presenter says <d>[English] Fast H3 runs on Apple silicon.</d>' \
  --height 480 --width 832 --num-frames 124 --steps 4 --seed 2026 \
  --output-path ./outputs/fasth3_int8.mp4

The MLX runtime loads one heavyweight component at a time. The source model's MiniMax H3 Community License applies to this converted checkpoint.

Validation and provenance

The conversion used FastVideo commit cf6a00b9be4675602126d6aeab902ad9a74810ea with MLX 0.32.2 on an Apple M4 Max. FastVideo verified all 13 source transformer shards, opened the safetensors header, validated 1,464 tensors and the quantization manifest, loaded the checkpoint, and completed a 124-frame 832x480 generation with the full H3 VAE.