---
title: mamba2-370m-8bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/mamba2-370m-8bit"
md_url: "https://www.modelscope.cn/models/mlx-community/mamba2-370m-8bit.md"
repository: mlx-community/mamba2-370m-8bit
last_updated: 2025-01-22
parameters: 104.1M
tensor_type:
  - F16
  - U32
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 974
stars: 0
---

# mamba2-370m-8bit

> mamba2-370m-8bit - mlx-community 在 ModelScope 开源的模型。mlx-community/mamba2-370m-hf-8bit

- **Repository**: mlx-community/mamba2-370m-8bit
- **Parameters**: 104.1M
- **Downloads**: 974
- **Stars**: 0
- **Last updated**: 2025-01-22

Source: https://www.modelscope.cn/models/mlx-community/mamba2-370m-8bit

---

# mlx-community/mamba2-370m-hf-8bit

The Model [mlx-community/mamba2-370m-hf-8bit](https://huggingface.co/mlx-community/mamba2-370m-hf-8bit) was
converted to MLX format from [rokyang/mamba2-370m-hf](https://huggingface.co/rokyang/mamba2-370m-hf)
using mlx-lm version **0.21.0**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/mamba2-370m-hf-8bit")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```