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

Multi-task Language Understanding on MMLU (Accuracy, Exit Position)

73.1Accuracy

Ministral3 8B

22.03635.29348.5561.807Apr 20, 2026May 1, 2026May 12, 2026May 23, 2026Jun 3, 2026Jun 14, 2026Jun 26, 2026
Updated 26d ago

Evaluation Results

MethodLinks
2026.04
73.1-
2026.04
72.72.01
2026.04
72.72.76
2026.04
66.7-
2026.04
65.44.48
2026.04
64.12.07
2026.06
59.58-
2026.06
57.83-
2026.06
56.62-
2026.06
56.4-
2026.06
55.92-
2026.06
55.8-
2026.06
55.72-
2026.06
55.34-
2026.06
52.96-
2026.06
50.93-
2026.06
50-
2026.06
48.83-
2026.06
47.01-
2026.06
38.8-
2026.06
30.1-
2026.06
29.7-
2026.06
29.1-
2026.06
25.7-
2026.06
25.4-
2026.06
24.7-
2026.06
24-