A 64GB Mac, and the local model that stopped needing a check (vickiboykis.com)
Her quality test isn't a benchmark. It's whether she has to verify the output against a paid model — and OpenAI's OSS-20B was the first where she did that a lot less often.
The thread's counter is blunt: boot Gemma 4 MoE expecting Claude Code behavior and you'll be very disappointed. Capable, but still far behind the frontier.
Sign in to save