Long-horizon software engineering on real codebases.
113 original software engineering tasks spanning 91 repositories and 5 languages. Each task requires large, novel fixes not sourced from existing public commits.
Model
Score
Opus 5Max
71.4%
Opus 5xHigh
70.2%
GPT-5.6 TerraMax
Kimi K3Max
66.4%
GPT-5.5xHigh
66.1%
GPT-5.6 LunaMax
64.3%
APEX NEWSLETTER
New benchmarks, leaderboard shifts, and research from the APEX team.
By subscribing you agree to receive updates from Mercor.Unsubscribe anytime.