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

MMLU Question Answering (Accuracy and Performance Gain)

78.7Accuracy

Mistral Small 24B Inst 2501

20.699235.757150.81565.8729Jan 30, 2026Feb 21, 2026Mar 16, 2026Apr 8, 2026May 1, 2026May 24, 2026Jun 16, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
78.7-0.6
2026.01
78.5-0.8
2026.06
77.8-
2026.01
76.7-1.5
2026.01
76.6-1.6
2026.06
76.6-
2026.06
75.1-
2026.06
73.04-
2026.01
65.3-0.1
2026.01
65.3-0.1
2026.01
63.9-0.8
2026.01
63.6-1.1
2026.01
62.6-1.3
2026.01
62.3-1.6
2026.06
61.27-
2026.06
61.08-
2026.06
60.4-
2026.06
60.34-
2026.01
60.21
2026.06
60-
2026.01
59.3-1
2026.01
58.7-1.3
2026.01
58.7-1.6
2026.01
58.7-0.4
2026.01
58.5-0.1
2026.01
58.5-1.5
2026.01
58.4-0.2
2026.06
58.29-
2026.06
57.55-
2026.06
56.68-
2026.06
56.19-
2026.06
55.96-
2026.01
55.9-1.1
2026.01
55.4-1.6
2026.06
55.39-
2026.06
53.82-
2026.06
52.73-
2026.06
50.67-
2026.01
45.90.1
2026.06
45.67-
2026.01
45.6-0.2
2026.01
43.81.5
2026.01
41.1-1.1
2026.06
31.52-
2026.06
22.95-
2026.06
22.93-