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

Multimodal Mathematical Reasoning on MathVerse mini

65.9Accuracy

Metis

17.43630.01842.655.182Mar 24, 2026Mar 26, 2026Mar 29, 2026Apr 1, 2026Apr 3, 2026Apr 6, 2026Apr 9, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.04
65.9
2026.04
61.3
2026.04
54.2
2026.04
52.7
2026.04
52.1
2026.04
50.3
2026.04
47.3
2026.04
45.6
2026.03
45.42
2026.03
45.24
2026.03
44.89
2026.03
44.44
2026.03
44.23
2026.03
42.09
2026.03
41.11
2026.03
40.54
2026.03
39.84
2026.04
39.8
2026.03
37.66
2026.03
23.36
2026.03
20.7
2026.04
19.3