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

Multitask Language Understanding on GlobalMMLU

73.3Accuracy

Marco-Mini-Instruct

13.583229.086644.5960.0934Dec 25, 2025Jan 24, 2026Feb 23, 2026Mar 25, 2026Apr 24, 2026May 24, 2026Jun 24, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
73.3--------------
2026.06
71.53-72.7-----6165.76-74.6672.682.45-
2026.06
71.12-72.18-----60.1565.68-74.3272.2582.12-
2026.06
70.73-71.72-----59.3965.6-73.9771.9181.8-
2026.04
70.2--------------
2026.06
69.3-70.84-----59.4664.68-72.6270.9-77.27
2026.04
69.2--------------
2026.06
69.02-70.53-----58.6564.48-72.5570.64-77.24
2026.06
68.78-70.23-----57.8864.28-72.6970.38-77.2
2026.06
68.63-70.04-----56.0963.58-71.5369.5281.03-
2026.04
67.4--------------
2026.06
66.74-68.57-----55.0962.13-69.8568.24-76.54
2026.06
59.28-64.86-----23.6657.46-68.1563.279.6576.97
2026.04
58.7--------------
2026.04
57--------------
2026.04
55.4--------------
2026.04
49--------------
2025.12
48.65--------------
2025.12
47--------------
2026.04
46.3--------------
2026.04
45.9--------------
2026.04
43--------------
2026.06
42.8-28.7-----45.6619.09-44.1438.2580.98-
2026.06
42.22-28.35-----45.5417.55-43.1537.8280.92-
2026.06
41.63-27.99-----45.4215.96-42.1437.3980.86-
2025.12
40.58--------------
2026.06
39.84-26.54-----43.212.45-40.5536.1280.15-
2025.12
39.27--------------
2026.04
38.4--------------
2025.12
36.04--------------
2025.12
26.1--------------
2026.06
15.88-12.78-----7.590.27-13.5811.8849.16-
2026.01
-27.940.531.132.533.832.830.432.938.931.1----
2026.01
-29.440.733.235.137.535.132.335.138.733.4----