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

Visual Mathematical Reasoning on MathVista (score)

81.9Score

GPT-5-Thinking

21.5837.2452.968.56Sep 30, 2025Oct 30, 2025Nov 30, 2025Dec 31, 2025Jan 31, 2026Mar 3, 2026Apr 3, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2025.09
81.9
2025.09
80.9
2025.09
75.6
2025.09
73.5
2026.04
70.1
2025.09
68.2
2026.04
68.2
2025.09
68
2026.03
67.9
2025.09
67.1
2026.03
66.3
2025.09
64.1
2025.09
62.3
2026.04
62.3
2026.04
60.5
2026.03
58.3
2026.03
57.3
2026.03
26.3
2026.03
23.9