LLM2014 Logic 2025-04: leaderboard
LLM2014 monthly logic benchmark for 2025-04, reporting aggregate logical-reasoning performance on the source score scale.
Metric: Median Score. Source: raw.githubusercontent.com. Saturation forecast: Estimated already saturated. 29 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 (High) | 85.85 | #124 (O3) |
| 2 | O4 Mini (High) | 80.87 | #180 (O4 Mini) |
| 3 | Gemini 2.5 Pro (Preview 03-25) | 77.84 | #123 |
| 4 | Grok 3 Mini | 72.63 | #281 |
| 5 | Doubao-1.5-Thinking-Pro | 67.94 | #134 |
| 6 | Claude 3.7 Sonnet (Thinking) | 67.34 | #249 (Claude 3.7 Sonnet) |
| 7 | Gemini 2.5 Flash (Thinking) | 63.69 | #245 (Gemini 2.5 Flash) |
| 8 | SenseNova V6 Reasoner | 55.28 | #306 |
| 9 | GPT-4.1 Mini | 54.55 | #355 |
| 10 | DeepSeek V3 (0324) | 50.91 | #340 |
| 11 | GPT-4.5 | 48.24 | #274 |
| 12 | Gemini 2.5 Flash | 47.61 | #245 |
| 13 | GPT-4.1 | 40.87 | #248 |
| 14 | Llama 4 Maverick | 34.74 | #464 |
| 15 | Llama 4 Scout | 16.16 | #656 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm2014-logic-2025-04 · How It Works · Data refreshed daily, snapshot 2026-10-20.