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

Code Generation on MBPP+ (Pass@1, Avg, ΔScore)

73.75Pass@1

R1-Oracle

48.498855.054461.6168.1656Mar 13, 2026Mar 25, 2026Apr 7, 2026Apr 19, 2026May 2, 2026May 14, 2026May 27, 2026
Updated 6d ago

Evaluation Results

MethodLinks
2026.03
73.7579.120
2026.05
72.6--
2026.05
72--
2026.05
71.6--
2026.05
70.9--
2026.05
70.7--
2026.05
70.6--
2026.05
70.2--
2026.05
69.8--
2026.05
69.6--
2026.05
68.8--
2026.05
68.3--
2026.05
67.72--
2026.05
67.5--
2026.05
67.5--
2026.05
66.75--
2026.05
65.3--
2026.03
6575.67-3.45
2026.03
63.7577.02-2.1
2026.03
56.2570.470
2026.03
5562.72-7.75
2026.03
5568.1-2.37
2026.05
54.5--
51.95--
51.95--
2026.05
51.9--
2026.05
51.59--
51.17--
2026.05
50.5--
50.39--
50.39--
2026.05
50.39--
2026.05
50.26--
50.26--
2026.05
50--
2026.05
50--
49.97--
49.61--
49.61--
2026.05
49.47--