OpenCodePapers

code-generation-on-programbench-golden-166-no-time-limit

Code IntelligenceCode Generation
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeAverage Test Pass RateModelNameReleaseDate
Introducing Claude Opus 5.5✓ Link91.2Claude Opus 5.5 (max effort) (mini-swe-agent)2026-09-22
Introducing Claude Fable 5.1 and Claude Mythos 5.187.6Claude Fable 5.1 (max effort) (mini-swe-agent)2026-09-01
Introducing Claude Haiku 5.582.0Claude Haiku 5.5 (max effort) (mini-swe-agent)2026-10-07
Introducing Claude Sonnet 5.5✓ Link79.7Claude Sonnet 5.5 (max effort) (mini-swe-agent)2026-09-28