Gemma-4 12B Coder — abliterated (weights, deprecated)

⚠️ Deprecated — do not use for new work. superseded by SFT v5 + abliterated — the same KL-guarded uncensoring applied on top of the SFT v5 tool-calling winner (no capability loss). Use that.

Replaced by tpls/gemma-4-12B-coder-fable5-composer2.5-v1-sft-v5-abliterated.

Uncensored gemma-4 12B coder weights (safetensors) — for fine-tuning, merging, or quantizing.

Ready-to-serve GGUF quants: tpls/gemma-4-12B-coder-fable5-composer2.5-v1-abliterated-GGUF.

⚠️ Tool-calling needs the recovery shim. The model emits gemma-4's native tool markup, which llama.cpp --jinja under-parses — wrap your endpoint with the tool-shim (see Tool-calling below) to get standard tool_calls.

At a glance

Use it — GGUF quantizations

Ready-to-serve GGUF quants live at tpls/gemma-4-12B-coder-fable5-composer2.5-v1-abliterated-GGUF (llama.cpp / Ollama one-liners on that card). These are the safetensors weights, for fine-tuning / merging / quantizing.

Tool-calling

Tool-calling works — but llama.cpp --jinja doesn't recognise gemma-4's native tool-call markup, so the bare parser under-reports calls. The model is fine; the parser is blind to the format. Recover standard tool_calls with a small serve-side post-processor (no weight change, no latency beyond a regex scan).

Ready-to-use → tpls/gemma4-tool-shim — a drop-in callback for OpenAI-compatible proxies, a standalone (dependency-free) example, and the pure parser, all Apache-2.0, with the full recovery algorithm documented. Point your OpenAI-compatible endpoint through it.

You send tools the usual OpenAI way (tools=[…]); the model emits native markup; the shim turns it into a standard tool_calls object:

# model completion (raw):
<|tool_call>get_weather{"city": "Paris", "units": "celsius"}

// after the shim:
{"finish_reason": "tool_calls",
 "message": {"role": "assistant", "content": null,
   "tool_calls": [{"id": "call_0", "type": "function",
     "function": {"name": "get_weather", "arguments": "{\"city\": \"Paris\", \"units\": \"celsius\"}"}}]}}

Intended use & limitations

Built for code generation and agentic tool use; serve locally via llama.cpp / Ollama, or use as a base to fine-tune / merge / quantize. Outputs can be wrong or fabricated — validate tool arguments before executing, and keep a human in the loop for anything consequential.

⚠️ Uncensored. For this variant the refusal direction has been ablated from the weights — safety guardrails are substantially removed and it will attempt requests a stock model would refuse. You are responsible for what you generate and how it's used; not suitable where refusal behaviour is itself a safety requirement.

Where this sits in the family

• base (upstream) — yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1
  • SFT v5 (weights)
    • SFT v5 + abliterated (weights)
      • SFT v5 + abliterated (GGUF)
    • SFT v5 (GGUF)

Provenance & reproduction

How this model was built — technique chain, training mix, and the exact knobs/pins, so the result is reproducible without any of our tooling.

Mechanics applied

1. abliteration

weight ablation degrades the canonical <|tool_call> token — the model tends to leak calls as text markup, so the native llama.cpp parser may not fire. See the tool-call recovery note below to get structured calls back.

Part of the Gemma-4 12B Coder — archive (superseded) collection.

Something not right, or a request? Open a discussion — happy to help.