MathArena - Project Euler 943-970: leaderboard
Project Euler computational math problems 943-970, evaluating models on exact-answer problem solving with mathematical insight and executable reasoning.
Metric: Accuracy (%, direct Project Euler problems 943-970). Source: matharena.ai. Saturation forecast: Around January 2027. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 (xHigh) | 87.5 | #79 (GPT-5.4) |
| 2 | Gemini 3.1 Pro (Preview) | 85.71 | #58 |
| 3 | Claude Opus 4.6 (High) | 83.93 | #63 (Claude Opus 4.6) |
| 4 | GPT-5.2 (High) | 80.36 | #110 (GPT-5.2) |
| 5 | Gemini 3.5 Flash | 78.57 | #60 |
| 6 | GLM-5.1 | 67.86 | #97 |
| 7 | GPT-5.1 (High) | 66.07 | #135 (GPT-5.1) |
| 8 | Gemini 3 Pro (Preview) | 64.29 | #66 |
| 9 | Kimi K2.5 (Thinking) | 63.39 | #144 (Kimi K2.5) |
| 10 | Gemini 3 Flash | 60.71 | #99 |
| 11 | GPT-5 (High) | 58.04 | #98 (GPT-5) |
| 12 | O4 Mini (High) | 48.21 | #180 (O4 Mini) |
| 13 | DeepSeek V3.2 (Thinking) | 47.32 | #205 (DeepSeek V3.2) |
| 14 | Grok 4 | 46.43 | #181 |
| 15 | Kimi K2 (Thinking) | 46.43 | #243 (Kimi K2) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=matharena-project-euler-943-970 · How It Works · Data refreshed daily, snapshot 2026-10-20.