Atlas

Benchmarks

← All benchmarks

GSO Opt@10

Code · 2025-05-29

GSO Opt@10 estimates the percentage of 102 repository-level optimization tasks where at least one of ten agent attempts both passes correctness tests and achieves at least 95% of the expert developer's speedup.

Top models (higher is better)

ModelScore
Claude 3.5 Sonnet (Oct 2024)15.7
O4 Mini12.7
Loading Atlas data…