MiniMax-H3

About

What MiniMax-H3 is

Instructions further below. GGUF for MiniMax-H3, compatible on most platforms including stablediffusion.cpp and Unsloth. Run MiniMax-H3 locally You can run MiniMax-H3 via Unsloth: https://github.com/unslothai/unsloth/ GGUF quantizations of MiniMaxAI/MiniMax-H3 MiniMax H3 is an omni-modal generative system that produces video with native stereo audio, up to 15 seconds at 24 FPS with 32 kHz stereo audio. Both halves of the runtime are in this repo: the denoisers and the Qwen3-VL text encoder they need. H3 ships two denoisers, and which one you load decides what the model can be given: - fl2vapruned, the H3-Base first-and-last-frame variant. Text, plus zero, one or two frames. - ref2vapruned, the reference variant. Text, plus reference pictures, videos and audio. They are separate checkpoints, not settings, so pick the one that matches the task. Both are quantized here at the same rungs, so a given quant costs about the same either way. Example UD-Q2KXL, the smallest rung here, at 960x544, 124 frames, 24 FPS, 8 steps, guidance 1.0, seed 11, on a single card. a red panda stepping along a mossy log in a misty forest, cinematic The GIF is downsampled and silent. For the full 960x544 clip with its native 32 kHz stereo audio track, play assets/h3ggufudq2kxl.mp4. H3 generates the audio jointly with the video, so the audio is part of the model output rather than something added afterwards. Files Text and frames, fl2vapruned: References, ref2vapruned: Text encoder, shared by both: The UD- rungs are dynamic, mixed-precision builds. The uniform rungs hold one type throughout. Pair the Q2KM text encoder with the two smallest denoisers and the Q4KM one with everything else. The text encoder and the VAEs are shared, so switching between the two denoisers costs one denoiser download and nothing else. The VAEs are not duplicated here, take them from Comfy-Org/MiniMax-H3. Three flags are not optional. --mode vidgen, or H3 takes the image path and aborts. Explicit --cfg-scale 1.0, because H3 is distilled and cfg-free and aborts above 1.0 while the default is 7.0. And --backend te=cpu, which keeps the 12 GB text encoder off the card. Add --offload-to-cpu to fit a smaller GPU. The pr

Summarised from the published model card. Read the full card on the HuggingFace links below.

Specifications

What it is

How it works

How video models work

Variants

Sizes & precisions

Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys unsloth-minimax-h3 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (unsloth-minimax-h3 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":"unsloth-minimax-h3","prompt":"a drone shot over a forest"}'

Details

Languages, data & research

Licence

Commercial licence needed

Sources

Weights & code

Explore

More video models