---
base_model: hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled
base_model_relation: finetune
library_name: transformers
pipeline_tag: image-text-to-text
license: apache-2.0
language:
- en
datasets:
- nohurry/Opus-4.6-Reasoning-3000x-filtered
- Jackrong/Qwen3.5-reasoning-700x
- Roman1111111/claude-opus-4.6-10000x
tags:
- transformers
- safetensors
- qwen
- qwen3.6
- qwen3_5_moe
- moe
- unsloth
- trl
- reasoning
- chain-of-thought
- conversational
- image-text-to-text
- text-generation-inference
- vllm
- mlx
- mlx-my-repo
model-index:
- name: Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU-Pro
      type: TIGER-Lab/MMLU-Pro
      split: test
    metrics:
    - type: exact_match
      value: 75.71
      name: exact_match, custom-extract, limited sample
---

# tkorsback/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-mlx-8Bit

The Model [tkorsback/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-mlx-8Bit](https://huggingface.co/tkorsback/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-mlx-8Bit) was converted to MLX format from [hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled](https://huggingface.co/hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled) 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("tkorsback/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-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)
```