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

Multimodal mathematical reasoning on MathVision (test)

39.3Accuracy

Qwen-VL-Max

17.9823.51529.0534.585Jan 7, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.01
39.3-
2026.01
37.9-
2026.01
31.2195.6
2026.01
30.6204.8
2026.01
30.4-
2026.01
30.2324.6
2026.01
29.9692.8
2026.01
29.6457.2
2026.01
27.1298.6
2026.01
26.8323.5
2026.01
25.6443
2026.01
25.2447.8
2026.01
24-
2026.01
23.4349.2
2026.01
21.2450.6
2026.01
20.1240.1
2026.01
18.8443