MathTutorBench — leaderboard

Open-ended math tutoring benchmark for measuring pedagogical capabilities such as mistake diagnosis, Socratic questioning, and scaffolding.

Metric: Overall Average (self-reported). Source: benchmarklist.com. Status: saturation imminent. 8 models tracked.

Top models

#ModelScore
1GPT-4o63.78
2Llama 3.1 70B Instruct55.22
3Llama 3.1 8B Instruct47
4Llama 3.2 3B Instruct46.89
5Qwen2.5-Math-7B-Instruct31.67

Interactive version: theaggregate.ai/benchmark?slug=mathtutorbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.