---
language:
- en
library_name: transformers
tags:
- glm
- MOE
- pruning
- compression
- mlx
- mlx-my-repo
license: mit
name: cerebras/GLM-4.6-REAP-252B-A32B
description: 'This model was obtained by uniformly pruning 30% of experts in GLM-4.6
  using the REAP method.

  '
readme: 'https://huggingface.co/cerebras/GLM-4.6-REAP-252B-A32B/main/README.md

  '
license_link: https://huggingface.co/zai-org/GLM-4.6/blob/main/LICENSE
pipeline_tag: text-generation
base_model: cerebras/GLM-4.6-REAP-252B-A32B
---

# saviochow/GLM-4.6-REAP-252B-A32B-mlx-2Bit

The Model [saviochow/GLM-4.6-REAP-252B-A32B-mlx-2Bit](https://huggingface.co/saviochow/GLM-4.6-REAP-252B-A32B-mlx-2Bit) was converted to MLX format from [cerebras/GLM-4.6-REAP-252B-A32B](https://huggingface.co/cerebras/GLM-4.6-REAP-252B-A32B) using mlx-lm version **0.26.4**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("saviochow/GLM-4.6-REAP-252B-A32B-mlx-2Bit")

prompt="hello"

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

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