README.md

---
license: apache-2.0
base_model: allenai/Olmo-3-32B-Think
language:
- en
datasets:
- allenai/Dolci-Think-RL
library_name: transformers
tags:
- mlx
---

# garrison/Olmo-3-32B-Think-mlx-2Bit

The Model [garrison/Olmo-3-32B-Think-mlx-2Bit](https://huggingface.co/garrison/Olmo-3-32B-Think-mlx-2Bit) was converted to MLX format from [allenai/Olmo-3-32B-Think](https://huggingface.co/allenai/Olmo-3-32B-Think) using mlx-lm version **0.28.3**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("garrison/Olmo-3-32B-Think-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)
```