---
title: gemma-4-31b-it-nvfp4
canonical_url: "https://www.modelscope.cn/models/mlx-community/gemma-4-31b-it-nvfp4"
md_url: "https://www.modelscope.cn/models/mlx-community/gemma-4-31b-it-nvfp4.md"
repository: mlx-community/gemma-4-31b-it-nvfp4
last_updated: 2026-07-06
license: gemma
model_type:
  - gemma4
architectures:
  - Gemma4ForConditionalGeneration
base_model:
  - google/gemma-4-31B-it
base_model_relation: quantized
parameters: 8.2B
tensor_type:
  - BF16
  - U8
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 290
stars: 2
tags:
  - mlx
  - mlx-vlm
  - gemma4
  - image-text-to-text
---

# gemma-4-31b-it-nvfp4

> gemma-4-31b-it-nvfp4 - mlx-community 在 ModelScope 开源的模型。mlx-community/gemma-4-31b-it-nvfp4

mlx-community/gemma-4-31b-it-nvfp4 是 ModelScope 魔搭社区上的 8.2B 参数机器学习模型，采用 gemma 许可，基于 google/gemma-4-31B-it 构建。

- **Repository**: mlx-community/gemma-4-31b-it-nvfp4
- **License**: gemma
- **Parameters**: 8.2B
- **Base model**: google/gemma-4-31B-it
- **Tags**: mlx, mlx-vlm, gemma4, image-text-to-text
- **Downloads**: 290
- **Stars**: 2
- **Last updated**: 2026-07-06

Source: https://www.modelscope.cn/models/mlx-community/gemma-4-31b-it-nvfp4

---

# mlx-community/gemma-4-31b-it-nvfp4

MLX conversion of [google/gemma-4-31B-it](https://huggingface.co/google/gemma-4-31B-it) for Apple silicon.

- Source revision: `3548789868c5356dbf307c98e6f609007b82b1e06f`
- Variant: `nvfp4`
- Converted with `mlx_vlm.convert` from the local `mlx-vlm` checkout.

## Usage

```bash
pip install mlx-vlm
python -m mlx_vlm.generate --model mlx-community/gemma-4-31b-it-nvfp4 --prompt "Describe this image." --image path/to/image.jpg
```