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

Instruction Following on CodaSet ID IFEVAL (test)

88.76Accuracy

DeepSeek-V3.1

76.560879.727982.89586.0621May 25, 2026
Updated 8d ago

Evaluation Results

MethodLinks
88.764.9
87.563.1
2026.05
87.564.5
2026.05
87.323
86.843.3
2026.05
86.843.2
86.62.4
2026.05
85.415
2026.05
85.413.5
85.1712.8
84.691.4
2026.05
84.193.6
82.933
2026.05
82.783.3
81.821.2
77.031