---
title: MiniMax-M3-4bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/MiniMax-M3-4bit"
md_url: "https://www.modelscope.cn/models/mlx-community/MiniMax-M3-4bit.md"
repository: mlx-community/MiniMax-M3-4bit
last_updated: 2026-06-20
license: other
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
model_type:
  - minimax_m3_vl
architectures:
  - MiniMaxM3SparseForConditionalGeneration
base_model:
  - MiniMaxAI/MiniMax-M3
base_model_relation: quantized
parameters: 67.5B
tensor_type:
  - BF16
  - U32
  - F32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 420
stars: 1
tags:
  - multimodal
  - moe
  - agent
  - coding
  - video
  - mlx
---

# MiniMax-M3-4bit

> MiniMax-M3-4bit - mlx-community 在 ModelScope 开源的模型。mlx-community/MiniMax-M3-4bit

mlx-community/MiniMax-M3-4bit 是 ModelScope 魔搭社区上的 67.5B 参数image-text-to-text模型，采用 other 许可，基于 MiniMaxAI/MiniMax-M3 构建。

- **Repository**: mlx-community/MiniMax-M3-4bit
- **License**: other
- **Tasks**: image-text-to-text
- **Parameters**: 67.5B
- **Base model**: MiniMaxAI/MiniMax-M3
- **Tags**: multimodal, moe, agent, coding, video, mlx
- **Downloads**: 420
- **Stars**: 1
- **Last updated**: 2026-06-20

Source: https://www.modelscope.cn/models/mlx-community/MiniMax-M3-4bit

---

# mlx-community/MiniMax-M3-4bit

This model was converted to MLX format from [`MiniMaxAI/MiniMax-M3`](https://huggingface.co/MiniMaxAI/MiniMax-M3)
using mlx-vlm version **0.6.3**.
Refer to the [original model card](https://huggingface.co/MiniMaxAI/MiniMax-M3) for more details on the model.

## Use with mlx

```bash
pip install -U mlx-vlm
```

```bash
python -m mlx_vlm.generate --model mlx-community/MiniMax-M3-4bit --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>
```