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

Code Generation on BigCodeBench (30% held-out tasks)

55.8Success Rate (SR)

APEX-EM (A5: Opus judge)

45.50448.17750.8553.523Mar 31, 2026
Updated 17d ago

Evaluation Results

MethodLinks
2026.03
55.89.9
2026.03
53.59.9
2026.03
51.55.6
2026.03
51.25.3
2026.03
50.82.3
2026.03
50.64.7
2026.03
50.34.4
2026.03
501.5
2026.03
49.40.9
2026.03
48.82.9
2026.03
48.5-
2026.03
48.50
2026.03
47.90.6
2026.03
46.50.6
2026.03
45.9-