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

Code Generation on BigCodeBench (Accuracy, Average Added Skill Tokens)

52.19Accuracy

Full Text

-1.99412.07326.1440.207Jun 10, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
52.194,628
2026.06
50.351,889
2026.06
49.34,676
2026.06
49.33,316
2026.06
49.121,414
2026.06
47.462,615
2026.06
46.933,172
2026.06
46.58708
2026.06
46.053,362
2026.06
45.531,414
2026.06
45.261,284
2026.06
44.21708
2026.06
43.770
2026.06
43.62,655
2026.06
40.960
2026.06
40.71,300
2026.06
29.47256
2026.06
0.88256
2026.06
0.18256
2026.06
0.09256