---
language:
- en
- zh
license: apache-2.0
tags:
- unsloth
- heretic
- uncensored
- abliterated
- fine tune
- creative
- creative writing
- fiction writing
- plot generation
- sub-plot generation
- fiction writing
- story generation
- scene continue
- storytelling
- fiction story
- science fiction
- romance
- all genres
- story
- writing
- vivid prosing
- vivid writing
- fiction
- roleplaying
- bfloat16
- all use cases
library_name: transformers
pipeline_tag: image-text-to-text
base_model:
- DavidAU/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking
---

[uploading...]

<small><font color="red">Ultimate:</font> Exceeds Qwen 3.6 27B performance, uncensored and NEO-Di-Matrix quants to bring that power in quant form.</small>

<h2>Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF</h2>

Freedom, a stronger model, and ultimate GGUF quant performance.

<B>Ultimate Power:</B>
- Heretic'ed and de-censored: The "nanny" was "evicted".
- Finetune via Unsloth post Heretic, and model exceeds root Qwen 3.6 27B (censored) model performance.
- NEO/NEO Code Di-matrix GGUF performance; with Q4ks clocking in at 94% of BF16/Full precision performance.

<B>Pure FREEDOM (Heretic stats):</B>

```
Metric 			This model 		Original model (Qwen/Qwen3.6-27B)
KL divergence 	0.0469 			0 (by definition)
Refusals 		4/100 			99/100
```

KLD: Less than .3 is great, lower than this is excellent. This is a measurement of how much the "heretic" version differs from "original model".

<B>STRONGER then Qwen 3.6 27B:</B>

Low level fine tune (post Heretic'ing) to boost the model's core power just a wee bit, don't want to mess with the "Qwen magic":

```
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]

BASE UNTUNED MODEL:

Qwen3.6-27B (by Qwen) [instruct mode]
mxfp8    0.647,0.803,0.910,0.773,0.450,0.806,0.742

```
NOTE: Instruct mode will often test higher than "thinking" mode due to token usage in thinking and context limits.

<B>Killer Quants:</B>

IQ2_M: 83% accuracy of BF16/full precision model at only 20% of the org model's size.

Q4K_S: 94% accuracy of BF16/full precision model at only 25% of the org model's size.

[ more to come ]