IneqMath — leaderboard
Olympiad-level inequality proof benchmark evaluating both final-answer correctness and step-wise reasoning soundness for chat and reasoning LLMs.
Metric: Overall Accuracy (self-reported). Source: benchmarklist.com. Status: saturation imminent. 42 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 (Medium) | 47 |
| 2 | Gemini 2.5 Pro (Preview 06-05) | 46 |
| 3 | Gemini 2.5 Pro | 43.5 |
| 4 | O3 (Medium) | 37 |
| 5 | GPT-5 Mini (Medium) | 30.5 |
| 6 | Gemini 2.5 Flash | 23.5 |
| 7 | GPT-OSS-120B | 23.5 |
| 8 | DeepSeek V3.1 | 15.5 |
| 9 | O4 Mini (Medium) | 15.5 |
| 10 | DeepSeek R1 0528 | 9.5 |
| 11 | O3 Mini (Medium) | 9.5 |
| 12 | Grok 4 | 8 |
| 13 | O1 (Medium) | 8 |
| 14 | DeepSeek V3 (0324) | 7 |
| 15 | Qwen 3 235B A22B | 6 |
Interactive version: theaggregate.ai/benchmark?slug=ineqmath · How the rankings work · Data refreshed daily, snapshot 2026-07-22.