Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Visual Mathematical Reasoning on DynaMath

60.9Accuracy

GPT-5

18.7829.71540.6551.585Jan 5, 2026Jan 11, 2026Jan 17, 2026Jan 23, 2026Jan 29, 2026Feb 4, 2026Feb 11, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.01
60.9
2026.02
59.04
2026.02
58.98
2026.02
58.7
2026.02
58.62
2026.02
58.4
2026.02
58.26
2026.02
58.06
56.3
2026.01
53.9
2026.02
51.24
2026.02
50.02
2026.02
46.33
2026.01
46.2
44.9
2026.01
42.5
39.7
2026.01
37.7
2026.01
37.3
2026.01
25.5
2026.01
21
2026.01
20.4