Applications

Claude Code ollama launch claude --model richardyoung/gemma-4-12b-coder-abliterated

Codex App ollama launch codex-app --model richardyoung/gemma-4-12b-coder-abliterated

OpenClaw ollama launch openclaw --model richardyoung/gemma-4-12b-coder-abliterated

Hermes Agent ollama launch hermes --model richardyoung/gemma-4-12b-coder-abliterated

Codex ollama launch codex --model richardyoung/gemma-4-12b-coder-abliterated

OpenCode ollama launch opencode --model richardyoung/gemma-4-12b-coder-abliterated

Gemma-4-12B-Coder-Abliterated

Abliterated build of gemma-4-12B-coder, a Gemma-4 12B coding model with refusals dramatically reduced via the Heretic library.

🚀 Overview

Abliterated version of yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1, a Gemma-4 12B model fine-tuned for verified-chain-of-thought Python coding. Refusal behavior was reduced with Heretic. The base refused nearly every “unsafe” prompt; this build engages with a far wider range.

📊 Abliteration Results

Metric	Before	After
Refusals (harmful-prompt eval)	¹⁰⁰⁄₁₀₀	⁷⁄₁₀₀
Reduction	–	93%
KL Divergence	–	0.058

A 93% refusal reduction at low KL (0.058), among the strongest reductions in this collection (the base was extremely refusal-heavy).

🎯 Key Features

Strong coding, verified chain-of-thought Python focus (Gemma-4 12B base)
93% fewer refusals via Heretic, low KL
5 GGUF quant tiers

🏷️ Available Versions

Tag	Size	BPW	Notes
IQ3_M	~5.5 GB	3.66	Smallest
IQ4_XS	~6.5 GB	4.25	Great quality/size
latest / Q4_K_M	~7.5 GB	4.85	Recommended
Q5_K_M	~8.7 GB	5.68	Higher quality
Q8_0	~13 GB	8.5	Near-lossless

💻 Quick Start

ollama run richardyoung/gemma-4-12b-coder-abliterated

⚠️ Format note: this finetune uses a custom <channel|>-based output format (the composer2.5 scheme). The packaged Modelfile uses the standard Gemma <start_of_turn> template; for best results you may need to adapt the chat template to the channel format, or call the API with raw: true and the model’s native format.

🛠️ Use Cases

Python code generation, chain-of-thought coding
Research on coding-model refusal behavior

🔧 Technical Details

Base Model: yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1 (Gemma-4 12B, gemma4)
Abliteration: Heretic, Trial 188 (100→7 refusals, KL 0.058)
Quantization: GGUF via llama.cpp (text generation)

⚠️ Disclaimer

This model has reduced safety guardrails and will engage with a wider range of prompts, including generating code that a stock model would refuse. Use responsibly and in accordance with applicable laws.

🙏 Acknowledgments

Base Model: yuxinlu1 (gemma-4-12B-coder) / Google (Gemma 4)
Abliteration: Heretic by p-e-w
Quantization: llama.cpp

Built & maintained by Richard Young · DeepNeuro

a Gemma-4 12B coding model with refusals dramatically reduced via the Heretic library.