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

Mathematical Reasoning on MathVision (Acc@1, Acc@4)

72Top-1 Accuracy

GPT-5-Thinking*

21.24834.42447.660.776Apr 1, 2026
Updated 17d ago

Evaluation Results

MethodLinks
2026.04
7279.2
2026.04
69.172.5
2026.04
31.339.7
2026.04
29.934.1
2026.04
28.232.9
2026.04
27.835.4
2026.04
26.430.3
2026.04
25.629.4
2026.04
23.241.6