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

Instruction Following on IFBench prompt (Accuracy)

73.4Accuracy

Super

40.01648.68357.3566.017May 8, 2026May 17, 2026May 27, 2026Jun 6, 2026Jun 15, 2026Jun 25, 2026Jul 5, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.07
73.4
2026.07
71.9
2026.07
71.4
2026.07
71.3
2026.05
70.82
2026.05
70.58
2026.05
67.43
2026.05
64.03
43.6
2026.05
43.28
2026.06
41.3