PutnamBench — leaderboard

PutnamBench: Measures mathematical reasoning, symbolic problem solving, proof construction, or competition-style problem solving.

Metric: Problems Solved. Source: github.com. Status: saturation imminent. 41 models tracked.

Top models

#ModelScore
1GPT-528
2internlm-7B4
3GPT-4o3
4Gemini 2.5 Pro (Preview 03-25)3
5O4 Mini (High)2
6DeepSeek R11
7GPT-4o Mini0
8Claude 3.7 Sonnet0
9DeepSeek V3 (0324)0
10Grok 3 Mini0
11O3 Mini0

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