evals.report
BenchmarksLabsCompareRun guidesIn the wild
Models
1 selected
MiniMax M3MiniMax
Benchmarks
3 selected
SWE-bench VerifiedCodingDeepSWECodingSWE-bench ProCoding
BenchmarkMiniMax M3OpenMiniMax
SWE-bench Verified% resolved
DeepSWE% resolved13.3%
SWE-bench Pro% resolved59.0%
% resolved
13.3%
% resolved
59.0%

No aggregate score is calculated. Each row uses its benchmark’s own metric. Compare rows independently.