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

Gemma-4 12B Coder — SFT v5 (weights)

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-v5-GGUF`.

💡 Pick this for the best tool-calling (our gate winner). For an uncensored model, use SFT v5 + abliterated.

At a glance

Use it — GGUF quantizations

Ready-to-serve GGUF quants live at `tpls/gemma-4-12B-coder-fable5-composer2.5-v1-sft-v5-GGUF` (llama.cpp / Ollama one-liners on that card). These are the safetensors weights, for fine-tuning / merging / quantizing.

Tool-calling gate

Served the GGUF on llama.cpp (llama-server --jinja), prompted 7 tool-use cases + 1 no-tool abstain, scored whether a structured tool call was emitted. raw = llama.cpp native parse; shim = same outputs re-parsed for gemma-4 native markup. Tools folded into the prompt at eval time, matching training.

The rows are this model under two parse paths (raw and shim); the shim path is how it's served in production.

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) ← you are here
• —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. sft-qlora

• —tools_mode: mixed (xLAM schemas folded; conditional taught)

Training data & mix

Public sources; weights/row-caps are the exact balance.

Pinned revisions (byte-exact reproduction):

• —Agent-Ark/Toucan-1.5M (Kimi-K2) @ 0df3cf37f2abefb380370cfb02eabea2a35ae782
• —Nanbeige/ToolMind (graphsyndatasets/graphsyn.jsonl) @ 8020ed1c03c367e4eb720ac3828ab4b0b95d8baf
• —NousResearch/hermes-function-calling-v1 (func-calling.json) @ dae3e1d28cfbcf4b915c04ea1e072030529b4bda
• —NousResearch/hermes-function-calling-v1 (func-calling-singleturn.json) @ dae3e1d28cfbcf4b915c04ea1e072030529b4bda
• —Salesforce/xlam-function-calling-60k (toolsmode=mixed, toolsratio=0.5 (schemas folded into ~half the prompts)) @ 26d14ebfe18b1f7b524bd39b404b50af5dc97866

Training hyperparameters

Training environment

Exact pins the run trained against (the base arch needs a recent transformers).

Part of the Gemma-4 12B Coder — active collection.

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