evals.report
BenchmarksLabsCompareRun guides
OpenAIo-series

o3

OpenAI · o-series. Released Apr 16, 2025.

6 results

Benchmark results 6

Compare this model
BenchmarkCategoryScoreMetricStatusDate
FrontierMathReasoning18.69%accuracyOfficialMay 30, 2026Details
Berkeley Function Calling LeaderboardTool use63.05%accuracyOfficialApr 12, 2026Details
ARC-AGI-2Reasoning6.53%accuracyOfficialMay 19, 2026Details
SWE-bench VerifiedCoding62.3%% resolvedOfficialMay 30, 2026Details
MMMU-ProMultimodal76.4%accuracyOfficialApr 8, 2026Details
SimpleQA VerifiedOther53.0%accuracyOfficialMay 30, 2026Details