Claude Opus 4.8 — 21.3% on BenchFae

Claude Opus 4.821.3% overall on BenchFae, ranked 4 of 15.

Category

Score

Automated

Fairy

Human

Full-Stack Web Apps

29%

26.8%

38%

22.8%

2D Game Development

25.6%

28.2%

28.8%

22.6%

3D Game Development

17.2%

19.2%

24%

13.8%

Website Frontend

19.8%

21.5%

23.7%

15.8%

Website Backend

28%

22.8%

34.4%

—

Debugging/Refactoring

29.8%

27.1%

33.2%

—

Adversarial Code Review

0%

0%

0%

—