evals.report
BenchmarksLabsCompareRun guides
AnthropicClaude Opus

Claude Opus 4.7

Anthropic · Claude Opus. Released Apr 16, 2026.

11 results

Benchmark results 11

Compare this model
BenchmarkCategoryScoreMetricStatusDate
FrontierMathReasoning43.79%accuracyOfficialMay 30, 2026Details
DeepSWECoding54.20%% resolvedOfficialDetails
ARC-AGI-3Reasoning0.18%accuracyOfficialMay 19, 2026Details
ARC-AGI-2Reasoning75.83%accuracyOfficialMay 19, 2026Details
LMArenaChat preference1480source-defined ratingOfficialMay 27, 2026Details
LiveBenchReasoning76.91%scoreOfficialJan 8, 2026Details
GPQA DiamondReasoning90.2%accuracyOfficialMay 30, 2026Details
SWE-bench VerifiedCoding83.5%% resolvedOfficialMay 30, 2026Details
Humanity's Last ExamReasoning39.04%accuracyOfficialMay 31, 2026Details
AIME (OTIS Mock)Reasoning97.8%accuracyOfficialMay 30, 2026Details
SimpleQA VerifiedOther50.6%accuracyOfficialMay 30, 2026Details