MiniMax-H3-Lora

About

What MiniMax-H3-Lora is

MiniMax-H3 Turbo LoRA — Diffusers Diffusers / PEFT conversion of larryvrh/MiniMax-H3-Turbo-Lora: a LoRA that lets MiniMax-H3 render joint video + synchronized stereo audio in about 4 sampling steps instead of the usual ~20. This repo ships: - converted LoRA weights in Diffusers PEFT layout (transformer..loraA/B.weight) - convert.py to turn the original ComfyUI / generate.py safetensors into that layout ⚠️ Early prototype. Same caveat as the upstream release: under-trained preview weights, not production quality. They already beat the base model at 4 steps (sharper detail, cleaner / better-synced audio), but treat this as a work-in-progress taste, not a finished product. Prefer the non-EMA ckpt500 weights by default. Weights Converted from the upstream Turbo LoRA (bf16, Weff = W + loraB @ loraA, alpha = rank so scale is 1). QKV is split into toq / tok / tov, and SwiGLU fc1 halves are swapped to match Diffusers' [value; gate] layout. Ranks: attention / MLP = 64, AdaLN = 16. Keys are prefixed with transformer. for MiniMaxH3Transformer3DModel.loadloraadapter. Requirements MiniMax-H3 is not in a released Diffusers build yet. Install Diffusers from main, plus PEFT: Base weights: Diffusers-format MiniMaxAI/MiniMax-H3 (or your local conversion). Use a non-pruned DiT; pruned time-conditioning layouts are not compatible with this LoRA (same restriction as upstream). Quick start (Diffusers) Diffusers already runs dual video / audio schedules (scheduler shift 12, audioscheduler shift 3). You do not need the ComfyUI Turbo custom sampler node; a wrong single-schedule sampler is what blows up audio at 4 steps in ComfyUI. Convert from the original Turbo LoRA Original weights live in larryvrh/MiniMax-H3-Turbo-Lora (ComfyUI module names, fused qkvproj / mlp.fc1). What convert.py does: 1. Renames ComfyUI paths onto MiniMaxH3Transformer3DModel (blocks. → transformerblocks., mlp.fc → ff.net., finallayer.adalnproj → normout.linear, …). 2. Splits fused attn.qkvproj LoRA into toq / tok / tov (shared A, row-split B in [qall; kall; vall] layout). 3. Swaps mlp.fc1 LoRA halves from [gate; value] to Diffusers SwiGLU [value; gate]. 4. Writes keys with a transformer. prefix for loadloraadapte

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

Details

Languages, data & research

Licence

Open weights

Sources

Weights & code

Explore

More video models