Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Human-Likeness Evaluation on COGCAPTCHA30 Combined (test)
Loading...
0
Fool Rate
Gemini 2.5 Pro
-3.16
18.17
39.5
60.83
May 7, 2026
Fool Rate
Mean P(Human)
Updated 26d ago
Evaluation Results
Method
Method
Links
Fool Rate
Mean P(Human)
Gemini 2.5 Pro
Training Status=In cla...
2026.05
0
0.15
Claude Sonnet
Training Status=In cla...
2026.05
0
0.02
GPT-5
Training Status=In cla...
2026.05
4
0.11
Base Qwen
Training Status=Held o...
2026.05
20
0.27
Centaur
Training Status=Held o...
2026.05
79
0.67
Feedback
Search any
task
Search any
task