Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

OJBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Code ReasoningOJBench
Accuracy10.34
24
Code GenerationOJBench ICPC 2025 (test)
Accuracy19.18
18
Competitive CodingOJBench
Best@8 Score45.9
16
CodingOJBench
Accuracy40.7
12
Code GenerationOJBench
Accuracy58.8
10
Algorithmic Correctness EvaluationOJBench
Overall Score71.5
3
Showing 6 of 6 rows