Skip to content
Longterm Wiki has not been actively maintained since May 2026. Some data is still updated occasionally, but please don't rely on this site being current or accurate.
Longterm Wiki

MathVista

Multimodal
A benchmark for mathematical reasoning in visual contexts — combines visual understanding with mathematical problem-solving across geometry, charts, scientific figures, and more.
Models Tested
4
Best Score
73.9%
Median Score
70.4%
Scoring: accuracy
Introduced: 2023-10
Maintainer: UCLA / Microsoft Research

Leaderboard (4 models)

#ModelDeveloperScore
🥇o1OpenAI
73.9%
🥈GPT-4.1 miniOpenAI
73.1
🥉Claude 3.5 SonnetAnthropic
67.7%
4GPT-4oOpenAI
63.8%