---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3.6-27B/blob/main/LICENSE
language:
- en
- zh
- ru
- es
- fr
- it
- ja
- ko
- af
- de
- ar
- tr
- is
- pl
- sw
- sv
- nl
- he
- id
- uk
- fa
- pa
- pt
- ms
- fi
- el
base_model:
- DavidAU/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking
library_name: mlx
tags:
- 7bit
- mixed-precision
- fine tune
- mtp
- kv quantization
- fused spda
- mlx
pipeline_tag: image-text-to-text
---
<div align="center"><img width="400px" src="https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3.6/logo.png"></div>

# Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking

***Added full MTP support, KV Quantization***
**Merged jinja chat template fix from froggeric/Qwen-Fixed-Chat-Templates**

## Instructions
You will need this branch of mlx-lm:
```
https://github.com/iamthecage/mlx-lm
```

Install mlx-lm from the cloned repo:
```
pip install -e . --no-build-isolation
```

Optional(highly recommended): 
Install OptiQ to enable KV Quantization and fused SPDA optimizations
```
pip install mlx-optiq
```

Serve with:
```
mlx_lm.server \                                 
  --model iamthecage/Qwen3.6-27B-Thinking-MLX-mixed-7.6bit-MTP \
  --mtp-draft --num-draft-tokens 4 \
  --kv-bits 8 --kv-group-size 64 --quantized-kv-start 2048 \
  --chat-template-args '{"enable_thinking": true, "preserve_thinking": true, "auto_disable_thinking_with_tools": true}' \
  --temp 0 --host 127.0.0.1 --port 8082
```

**MTP Enabled**
INFO - Metrics: prompt 23123 tok @ 318.8 tok/s | gen 157 tok @ 20.2 tok/s | peak 34.22 GB | mtp 116/157 accepted (74%) accept-len 3.83

**MTP Disabled**
INFO - Metrics: prompt 28152 tok @ 362.8 tok/s | gen 144 tok @ 9.8 tok/s | peak 34.22 GB

**Quality**: quantized (***mixed quants per tensor**, group size: 32, **7.652 bpw***)

*Most tensors use 6-bit or 8-bit affine quantization with a group size 32*.

Fully uncensored and then fine-tuned by DavidAU.

### Abliteration metrics
| Metric | This model | Original model ([unsloth/Qwen3.6-27B](https://huggingface.co/Qwen/Qwen3.6-27B)) |
| :----- | :--------: | :---------------------------: |
| **KL divergence** | 0.0469 | 0 *(by definition)* |
| **Refusals** | 4/100 | 99/100 |

```
IN HOUSE BENCHMARKS [by Nightmedia]:
         arc-c arc/e boolq hswag obkqa piqa  wino
Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking
mxfp8    0.673,0.846,0.905... [instruct mode]
Qwen3.6-27B-Heretic-Uncensored-Finetune-Thinking
mxfp8    0.669,0.835,0.906,...   [instruct mode]
---
BASE UNTUNED MODEL:
Qwen3.6-27B HERETIC (by llmfan46) [instruct mode]
mxfp8    0.644,0.788,0.902,...
Qwen3.6-27B (by Qwen) [instruct mode]
mxfp8    0.647,0.803,0.910,0.773,0.450,0.806,0.742
```

-----
### Source
This model was converted to MLX format from [`DavidAU/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking`](https://huggingface.co/DavidAU/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking) using mlx-vlm version **0.4.4**.