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
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 63.78 |
| 2 | Llama 3.1 70B Instruct | 55.22 |
| 3 | Llama 3.1 8B Instruct | 47 |
| 4 | Llama 3.2 3B Instruct | 46.89 |
| 5 | Qwen2.5-Math-7B-Instruct | 31.67 |
Interactive version: theaggregate.ai/benchmark?slug=mathtutorbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.