MathArena - IMProofBench Proofs — leaderboard
Metric: Accuracy (%). Source: matharena.ai. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 (High) | 49.56 |
| 2 | Gemini 2.5 Pro | 34.2 |
| 3 | O4 Mini (High) | 31.37 |
| 4 | Grok 4 | 28.98 |
| 5 | Claude Opus 4.1 (Thinking) | 24.62 |
Interactive version: theaggregate.ai/benchmark?slug=matharena-improofbench-proofs · How the rankings work · Data refreshed daily, snapshot 2026-07-22.