---
title: GLM-4.6V-mxfp4
canonical_url: "https://www.modelscope.cn/models/mlx-community/GLM-4.6V-mxfp4"
md_url: "https://www.modelscope.cn/models/mlx-community/GLM-4.6V-mxfp4.md"
repository: mlx-community/GLM-4.6V-mxfp4
last_updated: 2026-02-23
license: mit
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
model_type:
  - glm4v_moe
architectures:
  - Glm4vMoeForConditionalGeneration
base_model:
  - zai-org/GLM-4.6V
base_model_relation: quantized
parameters: 17.6B
tensor_type:
  - U8
  - BF16
  - U32
  - F32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - zh
  - en
inference_backends:
  - "deploy_task vlm"
  - "sglang 0.5.2"
downloads: 43
stars: 0
tags:
  - mlx
---

# GLM-4.6V-mxfp4

> GLM-4.6V-mxfp4 - mlx-community 在 ModelScope 开源的模型。mlx-community/GLM-4.6V-mxfp4 This model was converted to MLX format from zai-org/GLM-4.6V using mlx-vlm with MXFP4 support. Refer to the original model card for more details on the model. Use with mlx

mlx-community/GLM-4.6V-mxfp4 是 ModelScope 魔搭社区上的 17.6B 参数image-text-to-text模型，采用 mit 许可，基于 zai-org/GLM-4.6V 构建，可用 deploy_task vlm、sglang 0.5.2 部署。

- **Repository**: mlx-community/GLM-4.6V-mxfp4
- **License**: mit
- **Tasks**: image-text-to-text
- **Parameters**: 17.6B
- **Base model**: zai-org/GLM-4.6V
- **Inference backends**: deploy_task vlm, sglang 0.5.2
- **Tags**: mlx
- **Downloads**: 43
- **Stars**: 0
- **Last updated**: 2026-02-23

Source: https://www.modelscope.cn/models/mlx-community/GLM-4.6V-mxfp4

---

# mlx-community/GLM-4.6V-mxfp4
This model was converted to MLX format from [`zai-org/GLM-4.6V`]() using [`mlx-vlm` with MXFP4 support](https://github.com/zhutao100/mlx-vlm.git).  
Refer to the [original model card](https://huggingface.co/zai-org/GLM-4.6V) for more details on the model.
## Use with mlx

```bash
pip install git+https://github.com/zhutao100/mlx-vlm.git
```

```bash
python -m mlx_vlm.generate --model mlx-community/GLM-4.6V-mxfp4 --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>
```