Compare / Claude Fable 5.1 vs MiniMax M3

Claude Fable 5.1 vs MiniMax M3

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

Claude Fable 5.1
Anthropic · profile
MiniMax M3
MiniMax · profile
Results compared
14
Shared sources
3
BenchmarkMetricClaude Fable 5.1MiniMax M3
APEX-AgentsreportedTasks passed% of tasks 68.6%37.7%
Arena (formerly LMArena)reportedConfirmed task successIPS effect estimate 0.17-0.14
Arena (formerly LMArena)reportedPraise over complaintIPS effect estimate 0.32-0.13
Arena (formerly LMArena)reportedSteerabilityIPS effect estimate 0.08-0.07
Arena (formerly LMArena)reportedTool groundingIPS effect estimate 0.00-0.00
Arena (formerly LMArena)reportedOverallArena rating 1,4931,433
Arena (formerly LMArena)reportedOverallArena rating 1,5011,434
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4851,450
Arena (formerly LMArena)reportedCreative writingArena rating 1,4841,406
Arena (formerly LMArena)reportedExpert promptsArena rating 1,5161,471
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4971,435
Arena (formerly LMArena)reportedOverallArena rating 1,5011,440
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4961,422
Vending-Bench 2reportedMoney after a yearUS dollars $5,421.56$2,157.77

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.