Claude Opus 5 — 23.5% on BenchFae

Claude Opus 523.5% overall on BenchFae, ranked 2 of 15.

Category

Score

Automated

Fairy

Human

Full-Stack Web Apps

29.7%

27%

39.6%

23.3%

2D Game Development

27.9%

32.1%

32.4%

23.5%

3D Game Development

20.3%

23.4%

28.6%

16%

Website Frontend

21.6%

22.8%

26.3%

17.6%

Website Backend

30.9%

26.6%

36.2%

—

Debugging/Refactoring

34.1%

33.1%

35.4%

—

Adversarial Code Review

0%

0%

0%

—