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

Code Generation on HumanEval (Accuracy, General Capability Average Accuracy)

35.4Accuracy

Pre-trained

28.6430.39532.1533.905Jun 1, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.06
35.450.95
2026.06
33.2749.51
2026.06
31.6648.32
2026.06
31.1648.05
2026.06
30.847.3
2026.06
30.5847.94
2026.06
30.4247.92
2026.06
29.9347.42
2026.06
28.944.2