---
license: apache-2.0
language:
- en
- zh
- ko
- ja
- multilingual
library_name: transformers
pipeline_tag: text-generation
tags:
- darwin
- darwin-reason
- reasoning
- advanced-reasoning
- chain-of-thought
- thinking
- reasoning-trace-distillation
- rtd
- darwin-delphi
- test-time-compute
- qwen3.6
- qwen
- gpqa
- benchmark
- open-source
- apache-2.0
- proto-agi
- vidraft
- eval-results
- mlx
- mlx-my-repo
base_model_relation: merge
base_model: FINAL-Bench/Darwin-28B-REASON
model-index:
- name: Darwin-28B-REASON
  results:
  - task:
      type: question-answering
      name: Question Answering
    dataset:
      name: GPQA Diamond
      type: Idavidrein/gpqa
      config: gpqa_diamond
    metrics:
    - type: accuracy
      value: 89.39
      name: Accuracy
---

# usermma/Darwin-28B-REASON-mlx-6Bit

The Model [usermma/Darwin-28B-REASON-mlx-6Bit](https://huggingface.co/usermma/Darwin-28B-REASON-mlx-6Bit) was converted to MLX format from [FINAL-Bench/Darwin-28B-REASON](https://huggingface.co/FINAL-Bench/Darwin-28B-REASON) using mlx-lm version **0.31.2**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("usermma/Darwin-28B-REASON-mlx-6Bit")

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)
```