README.md

---
license: mit
library_name: transformers
datasets:
- FractalAIResearch/Fathom-V0.4-SFT-Shortest-Chains
- FractalAIResearch/Fathom-V0.6-Iterative-Curriculum-Learning
base_model: FractalAIResearch/Fathom-R1-14B
tags:
- mlx
---

# cnfusion/Fathom-R1-14B-mlx-8Bit

The Model [cnfusion/Fathom-R1-14B-mlx-8Bit](https://huggingface.co/cnfusion/Fathom-R1-14B-mlx-8Bit) was converted to MLX format from [FractalAIResearch/Fathom-R1-14B](https://huggingface.co/FractalAIResearch/Fathom-R1-14B) using mlx-lm version **0.22.3**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("cnfusion/Fathom-R1-14B-mlx-8Bit")

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)
```