Anthropic Rank #1

claude-opus-4-5-20251101

Evaluated on January 14, 2026

Overall Score

30.5

33 / 33 tasks ran

Score Breakdown

20% weight

45.0

20 / 20 tasks ran

Tasks requiring < 1 hour

35% weight

40.0

10 / 10 tasks ran

Tasks requiring a few hours

45% weight

16.7

3 / 3 tasks ran

Tasks requiring > 1 day

Tasks ran = tasks succesfully attempted (was not refused by the provider's filters).