evals.report
BenchmarksLabsCompareRun guides
AnthropicClaude Opus

Claude Opus 4.5

Anthropic · Claude Opus. Released Nov 1, 2025.

8 results

Benchmark results 8

Compare this model
BenchmarkCategoryScoreMetricStatusDate
FrontierMathReasoning20.69%accuracyOfficialMay 30, 2026Details
Berkeley Function Calling LeaderboardTool use77.47%accuracyOfficialApr 12, 2026Details
LiveBenchReasoning75.96%scoreOfficialJan 8, 2026Details
SWE-bench ProCoding45.89%% resolvedOfficialMay 30, 2026Details
GPQA DiamondReasoning86.0%accuracyOfficialMay 30, 2026Details
SWE-bench VerifiedCoding76.7%% resolvedOfficialMay 30, 2026Details
Humanity's Last ExamReasoning25.8%accuracyOfficialMay 31, 2026Details
MMMU-ProMultimodal73.9%accuracyOfficialApr 8, 2026Details