Compare / Claude Haiku 4.5 vs Kimi K2 0711

Claude Haiku 4.5 vs Kimi K2 0711

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

Claude Haiku 4.5
Anthropic · profile
Kimi K2 0711
Moonshot AI · profile
Results compared
17
Shared sources
3
BenchmarkMetricClaude Haiku 4.5Kimi K2 0711
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4201,409
Arena (formerly LMArena)reportedCreative writingArena rating 1,3871,371
Arena (formerly LMArena)reportedExpert promptsArena rating 1,4521,417
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4151,381
Arena (formerly LMArena)reportedOverallArena rating 1,4141,418
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,3991,375
Berkeley Function Calling Leaderboard (BFCL) V4reportedIrrelevance detection% correct 95.3%87.3%
Berkeley Function Calling Leaderboard (BFCL) V4reportedMemory% correct 54.4%29.0%
Berkeley Function Calling Leaderboard (BFCL) V4reportedMulti-turn tasks% correct 53.6%50.6%
Berkeley Function Calling Leaderboard (BFCL) V4reportedOverall accuracy% correct 68.7%59.1%
Berkeley Function Calling Leaderboard (BFCL) V4reportedRelevance detection% correct 62.5%75.0%
Berkeley Function Calling Leaderboard (BFCL) V4reportedSingle-turn calls (curated)% correct 86.5%81.6%
Berkeley Function Calling Leaderboard (BFCL) V4reportedSingle-turn calls (user-contributed)% correct 78.7%78.7%
Berkeley Function Calling Leaderboard (BFCL) V4reportedWeb search% correct 83.5%66.5%
UGI LeaderboardreportedRequested-length error% off the requested word count 6.0%22.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.390.29
UGI LeaderboardreportedWriting scorescore out of 100 52.252.5

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.