Arena benchmark
Vision model ratings and results.
Updated Aug 11, 2026. Data as of 2026-08-11
Models are ordered by their Arena rank.
A closer view of the leading model ratings.
What Vision measures and how to read its results.
Vision reports rating values and currently defines 11 evaluation categories.
Arena results are preference or task-outcome signals. They are not automatically equivalent to capability benchmark scores.
Common questions about Vision.
Claude Fable 5 is currently ranked first.
This Arena reports rating values across 11 configured evaluation categories.
No. Arena results remain a separate preference or task-outcome signal.
100 model results are currently shown.