all the models — AI benchmark observatory
← Benchmarks

MathVerse

MathVerse evaluates multimodal mathematical reasoning, testing whether models genuinely interpret visual math diagrams rather than relying on text.

id mathverse · max 1 · 2 models reported

#ModelScore
1Seed 2.1 Pro
ByteDance
0.90
2Seed 2.1 Turbo
ByteDance
0.89
MathVerse Leaderboard · all the models