Buckets:
metadata
license: other
license_name: minimax-h3-community-license-agreement
license_link: https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/LICENSE
tags:
- comfyui
- diffusion-single-file
base_model:
- MiniMaxAI/MiniMax-H3
MiniMax H3
Repackaged model files for ComfyUI.
Original model repository: https://huggingface.co/MiniMaxAI/MiniMax-H3
The Qwen3-VL-32B nvfp4_awq quant is converted from: https://huggingface.co/cybermotaz/Qwen3-VL-32B-Instruct-NVFP4
This nvfp4 text encoder does not require Blackwell GPU to use.
For diffusion models prefer int8_convrot if you are able to use pytorch with cu130.
fp8_scaled should only be used if you cannot use int8_convrot.
Place the files in the following folders:
๐ ComfyUI/
โโโ ๐ models/
โ โโโ ๐ diffusion_models/
โ โ โโโ minimax_h3_fl2va_bf16.safetensors
โ โ โโโ minimax_h3_fl2va_int8_convrot.safetensors
โ โ โโโ minimax_h3_fl2va_pruned_bf16.safetensors
โ โ โโโ minimax_h3_fl2va_pruned_int8_convrot.safetensors
โ โ โโโ minimax_h3_fl2va_pruned_fp8_scaled.safetensors
โ โ โโโ minimax_h3_ref2va_bf16.safetensors
โ โ โโโ minimax_h3_ref2va_int8_convrot.safetensors
โ โ โโโ minimax_h3_ref2va_pruned_bf16.safetensors
โ โ โโโ minimax_h3_ref2va_pruned_int8_convrot.safetensors
โ โ โโโ minimax_h3_ref2va_pruned_fp8_scaled.safetensors
โ โโโ ๐ text_encoders/
โ โ โโโ qwen3vl_32b_minimax_h3_bf16.safetensors
โ โ โโโ qwen3vl_32b_minimax_h3_int8_convrot.safetensors
โ โ โโโ qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
โ โโโ ๐ vae/
โ โ โโโ minimax_h3_audio_vae_fp32.safetensors
โ โ โโโ minimax_h3_video_vae_fp16.safetensors
Workflows
Xet Storage Details
- Size:
- 2.17 kB
- Xet hash:
- 66f70e8771e54e4679fc8b9e4a339e2be1a8a2d400cf61a5790815bace1a93b0
ยท
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.