⚠️ Deprecated — do not use for new work. tool-calling regression — the reasoning-heavy mix shifted the model toward conversational answering; tool definitions were never in the prompt (ROADMAP §1). Fixed in v5.

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

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-sft-v4-GGUF.

At a glance


Type	Model weights (safetensors)
Techniques	—
Tool-calling	native `--jinja`
Status	⚠️ Deprecated → `tpls/gemma-4-12B-coder-fable5-composer2.5-v1-sft-v5`
Use	GGUF quants: `tpls/gemma-4-12B-coder-fable5-composer2.5-v1-sft-v4-GGUF`

Use it — GGUF quantizations

Ready-to-serve GGUF quants live at tpls/gemma-4-12B-coder-fable5-composer2.5-v1-sft-v4-GGUF (llama.cpp / Ollama one-liners on that card). These are the safetensors weights, for fine-tuning / merging / quantizing.

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.

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

Step	Technique	What it does	Provenance

Part of the Gemma-4 12B Coder — archive (superseded) collection.

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