Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Competitive Programming on CP-275 audited version 2
Loading...
80.7
Correctness (%)
AlgoSkill
8.836
27.493
46.15
64.807
Jun 29, 2026
Correctness (%)
Updated 25d ago
Evaluation Results
Method
Method
Links
Correctness (%)
AlgoSkill
Backbone=Claude 3 Haiku
2026.06
80.7
CoT
Backbone=Claude 3 Haiku
2026.06
79.3
Direct
Backbone=Claude 3 Haiku
2026.06
78.2
AlgoSkill
Backbone=Gemini-2.5-Flash
2026.06
18.2
CoT
Backbone=Gemini-2.5-Flash
2026.06
16
Direct
Backbone=Gemini-2.5-Flash
2026.06
11.6
Feedback
Search any
task
Search any
task