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

Language Understanding on MMLU (Accuracy and Forgetting)

87.56MMLU Accuracy

Qwen3-14B

49.423259.324169.22579.1259Nov 17, 2025Dec 18, 2025Jan 18, 2026Feb 18, 2026Mar 21, 2026Apr 21, 2026May 22, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.03
87.56-
2026.03
87.02-
2026.03
85.16-
2026.03
84.34-
2025.11
84.32-
2025.11
83.94-
2025.11
83.78-
2025.11
83.69-
2025.11
83.45-
2025.11
82.66-
2025.11
82.65-
2025.11
82.14-
2025.11
80.16-
2026.04
77-
2026.04
75-
2026.04
75-
2026.03
73.2-
2026.03
73.12-
2026.03
73.06-
2026.03
730.2
2026.03
72.96-
2026.03
72.96-
2026.03
72.94-
2026.03
71.91-
2026.03
71.88-
2026.03
71.82-
2026.04
71-
2026.03
70.72.5
2026.03
70.6-
2026.03
70.10.5
2026.05
70.1-
2026.05
70.1-
2026.03
700.6
2026.04
70-
2026.04
70-
2026.05
70-
2026.03
69.80.8
2026.05
69.7-
2026.05
69.4-
2026.05
69.2-
2026.03
68.62.8
2026.03
68.42.2
2026.03
68.2-
2026.03
68.13-
2026.03
68.12-
2026.03
68.12-
2026.03
68.02-
2026.03
66.46.8
2026.03
66.19-
2026.05
65.8-
2026.05
65.5-
2026.03
65.24-
2026.03
64.91-
2026.05
64.2-
2026.04
64.1-
2026.05
64.1-
2026.05
63.4-
2026.04
63.2-
2026.04
62.5-
2026.04
61.6-
2026.04
60.9-
2026.04
60.9-
2026.04
60.8-
2026.04
60.2-
2026.04
56.1-
2026.04
55.04-
2026.04
55.01-
2026.04
54.98-
2026.04
54.9-
2026.04
54.66-
2026.04
54.58-
2026.04
54.45-
2026.04
54.33-
2026.04
54.2-
2026.04
54.19-
2026.04
54-
2026.04
54-
2026.04
53.97-
2026.04
53.53-
2026.04
53.44-
2026.04
53.43-
2026.04
53.4-
2026.04
53.35-
2026.04
52.96-
2026.04
52.7-
2026.04
52.7-
2026.04
52.34-
2026.04
52.23-
2026.04
52.22-
2026.04
52.1-
2026.04
52-
2026.04
51.96-
2026.04
51.92-
2026.04
51.86-
2026.04
51.5-
2026.04
51.39-
2026.04
51.2-
2026.04
51.15-
2026.04
51.08-
2026.04
50.89-
Showing 100 of 132 rows