.@EVO__HQ just doubled Claude Code's accuracy on a hard benchmark In practice: Claude Code: 5/20 → 11/20 on a hard benchmark after 50-60 @EVO__HQ experiments Genome sequencing: 95% compute reduction with better results than baseline @alokbishoyi97 on what EVO does and how it works 👇 EVO looks

관련 프로젝트

Baseline