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

Code on HumanEval+

79.9Accuracy

SUN

0.65221.22641.862.374Jan 21, 2026Feb 10, 2026Mar 3, 2026Mar 24, 2026Apr 13, 2026May 4, 2026May 25, 2026
Updated 7d ago

Evaluation Results

MethodLinks
2026.03
79.9
2026.03
78
2026.03
76.2
2026.03
74.3
2026.01
73.2
2026.01
73.15
2026.01
72.48
2026.01
70.1
2026.01
69.8
2026.01
65.9
2026.01
64
2026.03
62.3
2026.01
62.2
2026.01
62.2
2026.01
61.6
2026.03
61.6
2026.03
59.8
2026.01
59.1
2026.03
58.5
2026.01
56.7
2026.01
56.7
2026.01
53.69
2026.01
53
2026.01
53
2026.01
51.68
2026.05
49.39
2026.05
46.95
2026.03
45.7
2026.05
45.12
2026.01
44.5
2026.01
44.5
2026.05
43.9
2026.05
43.29
2026.01
42.95
2026.03
41.5
2026.05
41.46
2026.03
39.6
2026.01
32.3
2026.03
29.9
2026.05
21.95
2026.05
21.95
2026.05
20.12
2026.01
3.7