evals.report
BenchmarksLabsCompareRun guidesIn the wild
AnthropicClaude Opus

Claude Opus 5

Anthropic · Claude Opus. Released Jul 24, 2026.

Claude Opus 5 is a model from Anthropic in the Claude Opus family, released Jul 24, 2026. evals.report tracks 8 reported Claude Opus 5 benchmark scores across AA-Omniscience: Knowledge and Hallucination Benchmark, ARC-AGI-1, ARC-AGI-2, ARC-AGI-3, BrowseComp, Humanity's Last Exam, DeepSWE, FrontierCode — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.

8 results

Benchmark results 8

Compare this model
BenchmarkCategoryScoreMetricStatusDate
AA-Omniscience: Knowledge and Hallucination BenchmarkReasoning37AA-Omniscience IndexOfficialJul 24, 2026Details
ARC-AGI-1Reasoning97.5%accuracyOfficialJul 24, 2026Details
ARC-AGI-2Reasoning90.42%accuracyOfficialJul 24, 2026Details
ARC-AGI-3Reasoning30.2%accuracyUnverifiedJul 24, 2026Details
BrowseCompAgents90.8%accuracyVerifiedJul 24, 2026Details
Humanity's Last ExamReasoning56.3%accuracyVerifiedJul 24, 2026Details
DeepSWECoding68.8%% resolvedVerifiedJul 24, 2026Details
FrontierCodeCoding53.4%weighted score (Main)OfficialJul 24, 2026Details