Compare / Claude Opus 4.8 vs Gemini 2.5 Pro

Claude Opus 4.8 vs Gemini 2.5 Pro

14 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.

Claude Opus 4.8
Anthropic · profile
Gemini 2.5 Pro
Google · profile
Results compared
14
Shared sources
3
BenchmarkMetricClaude Opus 4.8Gemini 2.5 Pro
Arena (formerly LMArena)reportedOverallArena rating 1,4861,431
Arena (formerly LMArena)reportedOverallArena rating 1,1951,142
Arena (formerly LMArena)reportedOverallArena rating 1,4691,462
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4901,437
Arena (formerly LMArena)reportedCreative writingArena rating 1,4671,443
Arena (formerly LMArena)reportedExpert promptsArena rating 1,5251,458
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4911,438
Arena (formerly LMArena)reportedOverallArena rating 1,4801,446
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4741,445
tau2-benchreportedConsistency (pass^4)% of tasks 22.7%1.0%
tau2-benchreportedTask success (pass^1)% of tasks 39.7%13.7%
UGI LeaderboardreportedRequested-length error% off the requested word count 6.0%19.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.400.40
UGI LeaderboardreportedWriting scorescore out of 100 65.965.2

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.