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

Coding on HumanEval (accuracy)

86.6Accuracy

ICaRus

25.03241.0165772.984Feb 27, 2026Mar 1, 2026Mar 3, 2026Mar 6, 2026Mar 8, 2026Mar 10, 2026Mar 13, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
86.6
2026.02
81.7
2026.02
68.3
2026.02
48.2
2026.02
48.2
2026.03
44.5
2026.03
44.5
2026.03
41.5
2026.03
40.2
2026.03
40.2
2026.03
39.6
2026.03
37.8
2026.03
37.8
2026.03
37.8
2026.03
37.8
2026.02
36.6
2026.03
35.3
2026.03
32.9
2026.03
32.3
2026.03
32.3
2026.03
31.1
2026.03
27.4