Oriel — AI & technology worth noticing, every day.

Comfy-Org/MiniMax-H3

AI model · https://huggingface.co/Comfy-Org/MiniMax-H3

First seen by Oriel on .

AI summary

This Hugging Face repository repackages MiniMax-H3 model files, quantizations, LoRAs, VAEs, embeddings, ControlNet patches, and workflows for ComfyUI.

  • Files include diffusion models, Qwen3-VL-32B text encoders, LoRAs, audio and video VAEs, ControlNet patches, and prompt embeddings.
  • The repository provides bf16, int8_convrot, w6a8, fp8_scaled, and nvfp4_awq variants across selected components.
  • For diffusion models, int8_convrot is preferred with PyTorch using cu130, while fp8_scaled is intended for systems unable to use it.
  • The nvfp4 text encoder does not require a Blackwell GPU, according to the repository instructions.
  • Included workflows cover image, text, reference, and multiframe video generation, image editing, text-to-image, and ControlNet use cases.

#comfyui #minimax-h3 #video generation #quantization

Generated by gpt-5.6-sol. Check the original for details.

Metric trends

Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained

Likes

Total likes reported by the source.

  • Hugging Face Models: 2,731, latest observation ; +48 over observed period since

Downloads (30d)

Rolling 30-day downloads. A change in this value is not new downloads during the observed period.

  • Hugging Face Models: 33,282,001, latest observation ; 0 over observed period since

Aliases

SourceExternal IDURL
Hugging Face ModelsComfy-Org/MiniMax-H3https://huggingface.co/Comfy-Org/MiniMax-H3 Read discussion