GGUF quantizations of igorls/gemma-4-12B-it-heretic, a fully-automatic decensored ("abliterated") version of google/gemma-4-12B-it produced with Heretic.

The decensored model has 0/100 genuine refusals on harmful prompts at a KL divergence of only 0.0284 from the original model — censorship removed with minimal loss of capability.

⚠️ Use non-thinking mode for best results

Gemma-4 is a hybrid thinking model, and the abliteration targets the direct (non-thinking) response — which is also Gemma-4's own default. For roleplay, creative writing, and the most reliable uncensored output, run with thinking disabled. In thinking mode the model produces good output too, but the chain of thought consumes the token budget and can leave the final answer truncated.

Runtime	How to disable thinking
Ollama (CLI)	`/set nothink` in the session
Ollama (API)	add `"think": false` to the request body
llama.cpp	omit `--jinja`, or use a prompt that closes the thought block
transformers	already non-thinking by default (`enable_thinking=False`)

If you do use thinking mode, set a large num_predict / num_ctx so the answer isn't cut off by the reasoning block.

Files

File	Quant	Size	Notes
`gemma-4-12B-it-heretic-Q4_K_M.gguf`	Q4_K_M	~7.4 GB	Recommended default. Runs on 8-12 GB VRAM.
`gemma-4-12B-it-heretic-Q8_0.gguf`	Q8_0	~12.7 GB	Near-lossless.

Usage

Ollama

bash

ollama run igorls/gemma-4-12B-it-heretic-GGUF
/set nothink                # recommended for roleplay / creative use

llama.cpp

llama-cli -m gemma-4-12B-it-heretic-Q4_K_M.gguf -p "Your prompt here"

Disclaimer

Safety alignment has been removed; this model will comply with requests the original refuses. You are responsible for your use of it and for complying with applicable laws and the base model's license.