Oriel — AI & technology worth noticing, every day.

FastVideo/FastH3-Omni-Comfy

AI model · https://huggingface.co/FastVideo/FastH3-Omni-Comfy

First seen by Oriel on .

AI summary

FastH3 Omni for ComfyUI provides BF16 and INT8 ConvRot diffusion models for generating video with synchronized audio from prompts and ordered image, video, and audio references.

  • Install one diffusion model, a text encoder, a video VAE, and the required audio VAE in the corresponding ComfyUI model directories.
  • Generation uses eight sampling steps with guidance 1.0, video shift 12.0, and audio shift 3.0.
  • Reference inputs are identified in prompts with <Picture 1>, <Video 1>, and <Audio 1> tags; the audio VAE is required for synchronized audio output.
  • The page requires a current ComfyUI build with native MiniMax-H3 support and notes that its outputs may differ from FastVideo and UniServe.
  • NVFP4 text encoding requires Blackwell hardware; the weights are governed by the MiniMax-H3 Community License Agreement.

#comfyui #video generation #audio generation #diffusion models

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: 176, latest observation ; +69 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: 324,335, latest observation ; +261,243 over observed period since

Aliases

SourceExternal IDURL
Hugging Face ModelsFastVideo/FastH3-Omni-Comfyhttps://huggingface.co/FastVideo/FastH3-Omni-Comfy Read discussion