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

Multitask Language Understanding on MMLU (val)

74.12Accuracy

GWT

21.163234.911648.6662.4084May 30, 2024Aug 31, 2024Dec 3, 2024Mar 7, 2025Jun 9, 2025Sep 11, 2025Dec 14, 2025
Updated 18d ago

Evaluation Results

MethodLinks
2025.01
74.1270.3883.8868.0677.14
2025.01
73.8570.4483.2367.876.9
2025.01
73.7770.5183.4367.3576.96
2025.01
73.7670.2183.4967.5276.87
2024.05
63.16----
2024.05
61.87----
2024.05
61.82----
2024.05
61.78----
2024.05
61.42----
2024.05
60.84----
2024.05
60.6----
2024.05
60.07----
2025.01
59.449.568.2556.2864.74
2024.05
59.38----
2025.01
59.3749.7368.755.9464.47
2025.01
59.3449.568.6156.5664.62
2025.01
59.3249.1768.1856.3264.71
2025.01
59.2349.8367.6356.5663.88
2024.05
57.69----
2024.05
57.29----
2024.05
57.15----
2024.05
57.01----
2024.05
56.52----
2024.05
55.03----
2024.05
54.89----
2025.01
54.5646.3964.8749.7359.38
2025.01
54.4646.5964.7449.7158.91
2025.01
54.2646.3664.5149.4258.91
2025.01
54.1746.4964.2849.3958.67
2025.01
53.4145.3663.4148.6358.36
2025.12
52.1----
2025.12
49.7----
2025.12
48----
2024.05
46.99----
2025.12
45.4----
2024.05
45.34----
2025.12
45.1----
2024.05
45.09----
2024.05
44.13----
2025.12
44.1----
2024.05
43.73----
2024.05
43.62----
2024.05
43.48----
2024.05
43.29----
2024.05
42.58----
2024.05
42.54----
2024.05
42.39----
2024.05
42.3----
2024.05
41.91----
2024.05
41.67----
2025.12
41.3----
2025.12
41.3----
2024.05
41.21----
2024.05
41.18----
2025.12
37.8----
2025.12
34----
2025.12
33----
2025.12
32.1----
2025.12
31.6----
2025.12
29.6----
2025.12
29.3----
2025.12
28.5----
2025.12
27.1----
2025.12
26.7----
2024.05
25.51----
2024.05
25.37----
2025.12
25.1----
2024.05
25----
2025.12
24----
2024.05
23.88----
2025.12
23.8----
2025.12
23.2----