Leaderboard

13 qualifying Arena votes across 2 comparisons. 33 candidate matchups entered; intervals are not yet computed.

#ModelEloDistributionRated matchups
01claude-fable-5156422
02gpt-5.5150823
03gemini14287

Head to head

Row wins against column, share of rated matchups.

claudegptgemini
claude-fable-563%100%
gpt-5.537%100%
gemini0%0%