Re:November 2026 Claude Fable 5.2 Scores 65 on Artificial An...
by eterm on Monday November 09, 2026 @04:54PM
(#50432804)
I saw one benchmark where Opus 5.5 max benched worse than opus 5.5 high, using "max" on any model seems foolish, but for opus 5.5 especially, where it seems to degrade at xhigh and max.
[ Parent
| on HN ]
There are no comments on this comment yet.