BREAKING
Claude Sonnet 4.6 Back in Focus
0
%
SWE-bench Verified
0
%
OSWorld computer use
OSWorld: 4.6 vs 4.5 vs Opus 4.6
Sonnet 4.5
61.4
Sonnet 4.6
72.5
Opus 4.6
72.7
0
%
over Sonnet 4.5
0
%
over Opus 4.5
Why 4.6 Slipped on Leaderboards
Rank Aside, Value Remains
AI NEWS BLITZ
Anthropic's Sonnet 4.6 is drawing fresh attention as its successor reshuffles the rankings.