LLM2014 Logic 2025-04 — leaderboard
Metric: Median Score. Source: raw.githubusercontent.com. 29 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 (High) | 85.85 |
| 2 | O4 Mini (High) | 80.87 |
| 3 | Gemini 2.5 Pro (03-25) | 77.84 |
| 4 | Grok 3 Mini | 72.63 |
| 5 | Claude 3.7 Sonnet (Thinking) | 67.34 |
| 6 | Gemini 2.5 Flash (Thinking) | 63.69 |
| 7 | GPT-4.1 Mini | 54.55 |
| 8 | DeepSeek V3 (0324) | 50.91 |
| 9 | GPT-4.5 | 48.24 |
| 10 | Gemini 2.5 Flash | 47.61 |
| 11 | GPT-4.1 | 40.87 |
| 12 | Llama 4 Maverick | 34.74 |
| 13 | Llama 4 Scout | 16.16 |
Interactive version: theaggregate.ai/benchmark?slug=llm2014-logic-2025-04 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.