---
title: MiniMax-H3-w4a8
canonical_url: "https://www.modelscope.cn/models/AI-ModelScope/MiniMax-H3-w4a8"
md_url: "https://www.modelscope.cn/models/AI-ModelScope/MiniMax-H3-w4a8.md"
repository: AI-ModelScope/MiniMax-H3-w4a8
last_updated: 2026-08-06
license: other
base_model:
  - MiniMaxAI/MiniMax-H3
base_model_relation: quantized
parameters: 11.7B
tensor_type:
  - F32
  - BF16
  - F8_E4M3
  - I8
  - F16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 2496
stars: 7
tags:
  - comfyui
  - safetensors
  - minimax-h3
  - w4a8
  - int4
  - int8
  - convrot
  - text-to-video
  - image-to-video
  - synchronized-audio-video
---

# MiniMax-H3-w4a8

> MiniMax-H3-w4a8 - AI-ModelScope 在 ModelScope 开源的模型。MiniMax H3 W4A8 for ComfyUI

AI-ModelScope/MiniMax-H3-w4a8 是 ModelScope 魔搭社区上的 11.7B 参数机器学习模型，采用 other 许可，基于 MiniMaxAI/MiniMax-H3 构建。

- **Repository**: AI-ModelScope/MiniMax-H3-w4a8
- **License**: other
- **Parameters**: 11.7B
- **Base model**: MiniMaxAI/MiniMax-H3
- **Tags**: comfyui, safetensors, minimax-h3, w4a8, int4, int8, convrot, text-to-video, image-to-video, synchronized-audio-video
- **Downloads**: 2496
- **Stars**: 7
- **Last updated**: 2026-08-06

Source: https://www.modelscope.cn/models/AI-ModelScope/MiniMax-H3-w4a8

---

# MiniMax H3 W4A8 for ComfyUI

Mixed W4A8 checkpoints for MiniMax H3 Ref2VA and FL2VA, converted from the pruned BF16 ComfyUI checkpoints.

> [!IMPORTANT]
> These checkpoints require the custom loader and comfy-kitchen W4A8 build from **[starsFriday/ComfyUI-W4A8-Loader](https://github.com/starsFriday/ComfyUI-W4A8-Loader)**. Do not use ComfyUI's standard **Load Diffusion Model** node; it does not register the experimental `asym_w4a8_int8` format.

## Files

| File | Variant | Size |
|---|---|---:|
| `minimax_h3_ref2va_pruned_w4a8_mixed.safetensors` | Ref2VA reference image, video, and audio workflows | 12.54 GB |
| `minimax_h3_fl2va_pruned_w4a8_mixed.safetensors` | FL2VA text/image and first/last-frame workflows | 12.54 GB |

## Installation

Stop ComfyUI, activate the Python environment used to run it, then install the node pack and native W4A8 implementation:

```bash
cd ComfyUI/custom_nodes
git clone https://github.com/starsFriday/ComfyUI-W4A8-Loader.git
cd ComfyUI-W4A8-Loader
python install_comfy_kitchen_w4a8.py
```

Download either checkpoint into one of these directories:

```text
ComfyUI/models/unet/
ComfyUI/models/diffusion_models/
```

Restart ComfyUI and use:

```text
advanced/loaders → Load Experimental W4A8 Diffusion Model
```

In an existing MiniMax H3 workflow, replace only the standard diffusion-model loader with this node. Keep the remaining VAE, text encoder, conditioning, sampler, and output connections unchanged.

## Requirements

- Linux and an NVIDIA GPU with compute capability SM 80 or newer
- Python 3.10 or newer
- A CUDA-enabled PyTorch build using CUDA 12.8 or newer
- CUDA Toolkit 12.8 or newer, including `nvcc`, to build comfy-kitchen
- A recent ComfyUI version; tested with ComfyUI 0.30.0, Python 3.12.9, PyTorch 2.9.1+cu130, CUDA Toolkit 13.1, and RTX 4090

Windows, AMD/ROCm, Apple Silicon, and CPU-only execution are not supported by the provided installer.

## Quantization

For each of the 50 MiniMax H3 transformer blocks, the converter quantizes these four weights:

- `attn.qkv_proj.weight`
- `attn.out_proj.weight`
- `mlp.fc1.weight`
- `mlp.fc2.weight`

The 200 targeted weights use packed 4-bit ConvRot storage, group size 16, ConvRot group size 256, a calibration-free 16-value codebook, and FP8 E4M3 scales. The optimized CUDA path quantizes activations to INT8 at runtime and uses INT8 GEMM. Non-targeted tensors remain in their source dtype, so these are mixed-precision checkpoints.

Each pruned BF16 source checkpoint is 40.23 GB; its mixed W4A8 result is 12.54 GB, a 68.8% file-size reduction. W4A8 is lossy, and file-size reduction does not guarantee an equal reduction in peak VRAM or workflow runtime.

## Source and license

- Base model: [MiniMaxAI/MiniMax-H3](https://huggingface.co/MiniMaxAI/MiniMax-H3)
- ComfyUI BF16 checkpoints: [Comfy-Org/MiniMax-H3](https://huggingface.co/Comfy-Org/MiniMax-H3)
- Loader, converter, installer, and usage documentation: [starsFriday/ComfyUI-W4A8-Loader](https://github.com/starsFriday/ComfyUI-W4A8-Loader)
- W4A8 native implementation: [Comfy-Org/comfy-kitchen](https://github.com/Comfy-Org/comfy-kitchen)

This is an independent experimental conversion, not an official MiniMaxAI or ComfyUI release. Use of the weights remains subject to the [MiniMax H3 Community License Agreement](https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/LICENSE).