Model reference · open weights
FastVideo-FastH3-4-step-Dense-DataFree-INT6 is an open-weight video model from FastVideo. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.
Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.
What it is
| Released by | FastVideo |
|---|---|
| Type | Video models |
| Task | Text→video |
| Runs with | fastvideo |
| Based on | FastVideo/FastVideo-FastH3-4-step-Preview-v1-Dense-DataFree |
| Released | 2026-09-01 |
| Popularity | 0 downloads / month |
| Licence | Commercial licence needed |
About
This repository contains the ready-to-run INT6 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 INT6 with group size 64. Activations
remain BF16. This export is dense-only and does not support --vsa.
| File | Purpose |
|---|---|
mlx_h3_dit.safetensors | Quantized H3 DiT weights, 15.41 GiB |
mlx_h3_dit.json | Architecture, quantization, and cached AdaLN schedule |
conversion_manifest.json | Source revision, converter version, validation, and checksums |
Weights SHA-256: fc585db6855f1209c288a9b593e617ec6ae9eecc7fde521abb622e244c742681
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-INT6 \
--local-dir ./FastH3-MLX-INT6
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-INT6 \
--prompt '(S1) A presenter says [English] Fast H3 runs on Apple silicon.' \
--height 480 --width 832 --num-frames 124 --steps 4 --seed 2026 \
--output-path ./outputs/fasth3_int6.mp4
The MLX runtime loads one heavyweight component at a time. The source model's MiniMax H3 Community License applies to this converted checkpoint.
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.
From the published model card. Full card on the HuggingFace links in the sidebar.
How it works
Using it via the API
Once AxForge deploys fastvideo-fasth3-4-step-dense-datafree-int6 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (fastvideo-fasth3-4-step-dense-datafree-int6 below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/videos/generations \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"fastvideo-fasth3-4-step-dense-datafree-int6","prompt":"a drone shot over a forest"}'
Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.