MiniMax H3 · 33B · video + audio · seven released variants

h3-rs

MiniMax H3 text/keyframe/reference-to-video-with-audio on your own desktop GPU, in Rust. One set of fused kernels runs through WebGPU (Vulkan, DX12, Metal) or CUDA; the CPU reference is the ground truth every kernel is checked against. Every recipe of the FastVideo MiniMax H3 cookbook is a --variant.

needs the local engine

Prompt Studio

Type a prompt, pick size, length and seed, and watch the denoising steps and the decode land, with the generated soundtrack. The model runs on your GPU; this page only talks to it.

Open the studio →

Variants

--variantWhat it isForwards
h3MiniMaxAI/MiniMax-H3 base: T2VA, FL2VA, Ref2VA (dense attention)49
fasth3-v1FastH3 4-step preview (VSA 0.9, tile 64)4
fasth3-v2FastH3 8-step V2 (VSA 0.8, video shift 10)8
fasth3-trimFastH3 Trim: 42 blocks, rank-16 AdaLN, NVFP4 linears, 26-block decoder8
fasth3-omniFastH3 Omni Ref2VA with Parallel Decoding Distillation (32 heads)8
pika-flexPika H3 Flex: FL2VA / Ref2VA LoRA, renoise sampler, 480p / 768p buckets6
regen-2kOpen-H3-Regenerate-2K: 768p draft → 2560×1440 (h3 regen2k)3

Run the engine

Download the base model once (VAEs, text encoder, tokenizer) and the variant's DiT, then start the engine and allow this site to reach it:

huggingface-cli download MiniMaxAI/MiniMax-H3 --include "vae/*" "audio_vae/*" "text_encoder/*" "tokenizer/*" --local-dir h3
huggingface-cli download FastVideo/FastVideo-FastH3-8-Step-V2 --local-dir fasth3-v2
h3 adaln-tables --variant fasth3-v2 --model-dir h3 --variant-dir fasth3-v2 --out v2-tables.safetensors
h3 serve --variant fasth3-v2 --model-dir h3 --variant-dir fasth3-v2 --tables v2-tables.safetensors \
    --weights int4-hqq-g64 --allow-origin https://h3-video.gialabs.sh

The engine listens only on 127.0.0.1 and answers this one site. Without --allow-origin it serves the studio itself at http://127.0.0.1:7860/.

Where it stands