Compare / Claude Opus 4.5 vs Kimi K3

Claude Opus 4.5 vs Kimi K3

15 results from 5 sources that measured both models. Each row is in the source's own units; there is no overall winner.

Claude Opus 4.5
Anthropic · profile
Kimi K3
Moonshot AI · profile
Results compared
15
Shared sources
5
BenchmarkMetricClaude Opus 4.5Kimi K3
Arena (formerly LMArena)reportedOverallArena rating 1,4591,472
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4721,489
Arena (formerly LMArena)reportedCreative writingArena rating 1,4691,458
Arena (formerly LMArena)reportedExpert promptsArena rating 1,5031,533
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4831,486
Arena (formerly LMArena)reportedOverallArena rating 1,4741,488
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4671,475
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 29.143.6
Artificial AnalysisreportedGPQA Diamond% of questions 86.6%93.5%
Artificial AnalysisreportedHumanity's Last Exam% of questions 30.1%46.9%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 77.3%88.7%
SimpleQA Verified (Epoch AI)reportedCorrect answers% of questions 45.7%50.6%
tau2-benchreportedConsistency (pass^4)% of tasks 11.3%17.5%
tau2-benchreportedTask success (pass^1)% of tasks 24.7%37.1%
Vending-Bench 2reportedMoney after a yearUS dollars $4,967.06$5,165.03

Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.