MI

FastVideo FastH3 Omni 8 Step V1 ComfyUI

MinimaxGGUF by vantagewithaiUpdated 2026-10-21
Download on Hugging Face
Parameters22.1B
Context—
Layers—
Architecturewan
Smallest quant10.9 GB
Quick answer

FastVideo FastH3 Omni 8 Step V1 ComfyUI needs about 15.9 GB of memory at Q4_K_M (a 13.7 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 4 quants, from 10.9 GB (Q3_K_M) to 19.2 GB (Q6_K).

FastVideo FastH3 Omni 8 Step V1 for ComfyUI – video generation with text and multimodal prompts

FastVideo FastH3 Omni 8 Step V1 is a 22.1 billion‑parameter model from the Minimax family. It generates video with synchronized audio from text, image, video, and audio references. The model is available in several GGUF quantizations for local inference.

Good at

  • Video generation with audio
  • Multimodal prompt handling
  • ComfyUI integration

Limits

  • No vision support in the quantized GGUF
  • Requires 8 k context for estimates
  • License is other

Best for

  • Creating short video clips from text prompts
  • Generating audio‑video pairs for creative projects
  • Testing multimodal generation pipelines

Who it’s for. Developers and researchers who need a local video generation model that accepts text and multimodal inputs. The model is suitable for experimentation with ComfyUI workflows and for building custom inference pipelines.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_M10.9 GB 12.7 GB
Too big Get file →
Q4_K_M13.7 GB 15.9 GB
Too big Get file →
Q5_K_M16.4 GB 18.8 GB
Too big Get file →
Q6_K19.2 GB 22.0 GB
Too big Get file →

For 8 GB VRAM use Q3_K_M (10.89 GB file, ~12.7 GB memory at 8k context). For 12 GB VRAM use Q4_K_M (13.71 GB file, ~15.9 GB memory). For 16 GB VRAM use Q5_K_M (16.36 GB file, ~18.8 GB memory). Larger VRAM allows Q6_K (19.17 GB file, ~22 GB memory).

What GPU do you need for FastVideo FastH3 Omni 8 Step V1 ComfyUI?

These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is Q4_K_M (13.7 GB file, about 15.9 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (19.2 GB file, about 22.0 GB in use).

Which FastVideo FastH3 Omni 8 Step V1 ComfyUI quant should you choose?

Q4_K_M is the usual starting point for FastVideo FastH3 Omni 8 Step V1 ComfyUI: it balances file size and quality well. Here it is a 13.7 GB file.

If memory is tight, the smallest file is Q3_K_M at 10.9 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturewan
Parameters22.1B

Source

Quantized byvantagewithai
FamilyMinimax
Licenseother
Downloads (HF)5,348
Last updated2026-10-21

Frequently asked questions about FastVideo FastH3 Omni 8 Step V1 ComfyUI

How much VRAM does FastVideo FastH3 Omni 8 Step V1 ComfyUI need?

About 15.9 GB at Q4_K_M with an 8K context window. That is the 13.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run FastVideo FastH3 Omni 8 Step V1 ComfyUI on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of FastVideo FastH3 Omni 8 Step V1 ComfyUI GGUF?

It depends on the quant: Q3_K_M is 10.9 GB, Q4_K_M is 13.7 GB, Q5_K_M is 16.4 GB, Q6_K is 19.2 GB.

Where can I download FastVideo FastH3 Omni 8 Step V1 ComfyUI GGUF?

The files are on Hugging Face in the vantagewithai/FastVideo-FastH3-Omni-8-Step-V1-ComfyUI-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does FastVideo FastH3 Omni 8 Step V1 ComfyUI use?

Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run FastVideo FastH3 Omni 8 Step V1 on 8 GB VRAM?

Yes, using the Q3_K_M quantization which requires about 12.7 GB of VRAM at 8k context.

Does the model support vision inputs?

The quantized GGUF does not include vision support; it focuses on text and multimodal prompts.

What is the context length for inference?

The memory estimates are based on an 8k context length; larger contexts will increase VRAM usage.

Other GGUFs of MiniMax-H3

Similar models

Browse all →

More from vantagewithai

All by vantagewithai →

More Minimax models

All Minimax →

Other medium models

All medium models →