Local models are good enough for agentic coding now

Vicki Boykis

Vicki Boykis reports that with Google's Gemma 4 family, local models achieve ~75% the accuracy/speed of frontier models for agentic coding workflows. She used gemma-4-26b-a4b via LM Studio to refactor Python code and write unit tests on a 2022 M2 Mac.

ModelsOllama

See 2 more sources

RT @vboykis: new post: how I develop recently using local models. the tooling is now good enough to...15 hours agoHugging Face

Local coding agents are good now, but only if you babysit them1 day agoBTA_LabsDiscuss

Vicki Boykis

15 hours ago