Studio models: MiniMax H3

Repackaged MiniMax H3 model files for studio.no.dev, which runs them in ComfyUI and uncomfy.

Original model repositories:

License

Every file here is a Model Derivative of MiniMax H3 and is distributed under the MiniMax H3 Community License Agreement with no additional terms. Video Rebirth distributes HyperFlow H3 Flash under that Agreement, and this repository redistributes it the same way. By downloading or using these files you accept that Agreement, including its territorial restrictions, commercial terms and Acceptable Use Policy. The upstream LICENSE governs.

NOTICE lists what changed in each file. This repository is not an official MiniMax or Video Rebirth product. Powered by MiniMax H3.

HyperFlow H3 Flash

Video Rebirth's two HyperFlow towers in ComfyUI's int8 layout. Both sample in 6 or 8 steps without PDD, and the int8 weights are copied unchanged from Video Rebirth's files.

File	Tower	Use it for
`minimax_h3_fl2va_hyperflow_flash_int8_convrot_comfy.safetensors` (16.3 GB)	fl2va: 3 blocks dropped, 102 linears low-rank	Text to video, and image to video from a first frame or first and last frames (`MiniMaxH3ImageToVideo`)
`minimax_h3_ref2va_hyperflow_flash_int8_convrot_comfy.safetensors` (16.8 GB)	ref2va: 1 block dropped, 114 linears low-rank	Video from reference images and audio (`MiniMaxH3ReferenceToVideo`)

They need ComfyUI-Studio-HyperFlow's HyperFlowInt8Loader and Video Rebirth's ComfyUI-HyperFlow-Flash for the schedule; ComfyUI's stock H3 loader won't load them. Use them with the stock MiniMax H3 text encoder and VAEs from Comfy-Org/MiniMax-H3.

📂 ComfyUI/
├── 📂 models/
│   ├── 📂 diffusion_models/
│   │   ├── minimax_h3_fl2va_hyperflow_flash_int8_convrot_comfy.safetensors
│   │   └── minimax_h3_ref2va_hyperflow_flash_int8_convrot_comfy.safetensors

Studio H3 V1

StudioH3-V1.safetensors (20.2 GB) is a MiniMax H3 checkpoint in ComfyUI's int8 layout (int8_tensorwise with convrot, group size 256). It loads with ComfyUI's stock UNETLoader (Load Diffusion Model). Use it with the stock MiniMax H3 text encoder and VAEs from Comfy-Org/MiniMax-H3.

📂 ComfyUI/
├── 📂 models/
│   ├── 📂 diffusion_models/
│   │   └── StudioH3-V1.safetensors

Downloads last month: 481

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for nodevorg/studio-models-h3

Base model

MiniMaxAI/MiniMax-H3

Finetuned

(227)

this model