evals.report
BenchmarksLabsCompareRun guides
AnthropicClaude Sonnet

Claude Sonnet 4.5

Anthropic · Claude Sonnet. Released Sep 29, 2025.

5 results

Benchmark results 5

Compare this model
BenchmarkCategoryScoreMetricStatusDate
Berkeley Function Calling LeaderboardTool use73.24%accuracyOfficialApr 12, 2026Details
LiveCodeBench ProCoding1412Codeforces EloOfficialMay 31, 2026Details
SWE-bench ProCoding43.60%% resolvedOfficialMay 30, 2026Details
SWE-bench VerifiedCoding71.3%% resolvedOfficialMay 30, 2026Details
MMMU-ProMultimodal68.9%accuracyOfficialApr 8, 2026Details