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

Mathematical Reasoning on MathVision (Accuracy)

83.9Accuracy

Qwen3.5

12.722431.201249.6868.1588May 21, 2025Jul 19, 2025Sep 17, 2025Nov 16, 2025Jan 15, 2026Mar 16, 2026May 15, 2026
Updated 16d ago

Evaluation Results

MethodLinks
2026.05
83.9
2026.05
83.4
2026.05
79.63
2026.05
78.9
2026.05
75.82
2026.05
74.6
2026.05
72
2026.05
68.95
2026.05
67.8
2026.05
65.7
2026.05
65.6
2026.05
64.7
2026.05
62.7
2026.05
58.5
2026.05
58.3
2026.05
58.2
2026.05
57.8
2026.05
57.1
2026.05
56.8
2026.05
54.8
2026.05
54.8
2026.05
54.6
2026.05
54.5
2026.05
54.1
2026.05
54
2026.05
53.8
2025.11
47.8
2025.11
41.3
2025.11
35.5
2025.11
35.2
2026.05
35
2025.11
34.9
2026.05
34.9
2026.05
34.8
2025.05
34.6
2026.05
34.5
2026.05
34.3
2026.05
34.1
2025.11
33.9
2025.11
33.2
2025.11
32.6
2025.11
31.5
2025.11
31.1
2025.05
30.9
2025.11
30.6
2025.11
30.3
2025.11
28.6
2025.05
27.7
2025.11
27.3
2026.05
25.66
2026.05
25.66
2025.11
25.5
2026.05
24.34
2026.05
24.34
2026.05
23.03
2026.05
22.69
2025.11
21.8
2026.05
21.38
2026.05
21.38
2026.05
20.72
2026.05
19.74
2026.05
18.75
2026.05
18.08
2025.11
17
2025.11
16.1
2026.05
15.46