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

Instruction Following Evaluation on FollowBench OOD

78.06HSR

Prompt

21.48436.17250.8665.548Jun 22, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
78.0684.62
2026.06
64.6677.12
2026.06
63.0174.7
2026.06
60.7374.9
2026.06
59.3372.4
2026.06
58.3170.61
2026.06
53.9266.2
2026.06
47.7664.09
2026.06
39.3461.59
2026.06
38.7656.31
2026.06
32.550.4
2026.06
31.8952.1
2026.06
23.8936.49
2026.06
23.6641.47