---
title: siglip-large-patch16-384-4bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/siglip-large-patch16-384-4bit"
md_url: "https://www.modelscope.cn/models/mlx-community/siglip-large-patch16-384-4bit.md"
repository: mlx-community/siglip-large-patch16-384-4bit
last_updated: 2025-03-30
license: apache-2.0
model_type:
  - siglip
architectures:
  - SiglipModel
parameters: 103.2M
tensor_type:
  - U32
  - F16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 454
stars: 0
tags:
  - vision
  - mlx
---

# siglip-large-patch16-384-4bit

> siglip-large-patch16-384-4bit - mlx-community 在 ModelScope 开源的模型。mlx-community/siglip-large-patch16-384-4bit

mlx-community/siglip-large-patch16-384-4bit 是 ModelScope 魔搭社区上的 103.2M 参数机器学习模型，采用 apache-2.0 许可。

- **Repository**: mlx-community/siglip-large-patch16-384-4bit
- **License**: apache-2.0
- **Parameters**: 103.2M
- **Tags**: vision, mlx
- **Downloads**: 454
- **Stars**: 0
- **Last updated**: 2025-03-30

Source: https://www.modelscope.cn/models/mlx-community/siglip-large-patch16-384-4bit

---

# mlx-community/siglip-large-patch16-384-4bit

The Model [mlx-community/siglip-large-patch16-384-4bit](https://huggingface.co/mlx-community/siglip-large-patch16-384-4bit) was converted to MLX format from [google/siglip-large-patch16-384](https://huggingface.co/google/siglip-large-patch16-384) using mlx-lm version **0.0.2**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/siglip-large-patch16-384-4bit")
response = generate(model, tokenizer, prompt="hello", verbose=True)
```