I fine tuned Gemma 4 12B for a 2.7x improvement on tool calling because I can't fit anything else comfortably into my 16 GBs of Vram

Gemma 12B is obviously a very well trained model, I always thought the fine tuning they did on it wasn't really cut out for agentic coding. From my own experiences it struggles to use the tools it's given from Github Copilot and is also very inept at the cli too. So I thought I'd kill two bird

中文全文翻译正在补齐，可先查看下方英文原文。

Gemma 12B is obviously a very well trained model, I always thought the fine tuning they did on it wasn't really cut out for agentic coding. From my own experiences it struggles to use the tools it's given from Github Copilot and is also very inept at the cli too. So I thought I'd kill two bird