LLM2014 Logic 2025-05: leaderboard
LLM2014 monthly logic benchmark for 2025-05, reporting aggregate logical-reasoning performance on the source score scale.
Metric: Median Score. Source: raw.githubusercontent.com. Saturation forecast: Estimated already saturated. 32 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 (High) | 85.14 | #124 (O3) |
| 2 | O4 Mini (High) | 80.16 | #180 (O4 Mini) |
| 3 | Grok 3 Mini | 69.06 | #281 |
| 4 | Gemini 2.5 Pro (Preview 05-06) | 68.33 | #147 |
| 5 | Claude Sonnet 4 (Thinking) | 68.22 | #199 (Claude Sonnet 4) |
| 6 | Qwen 3 235B A22B (Thinking) | 66.6 | #313 (Qwen 3 235B A22B) |
| 7 | Doubao-1.5-Thinking-Pro | 65.9 | #134 |
| 8 | Claude Opus 4 | 60.12 | #163 |
| 9 | GPT-4.1 Mini | 53.52 | #355 |
| 10 | SenseNova V6 Reasoner | 52.19 | #306 |
| 11 | Claude Sonnet 4 | 52.08 | #199 |
| 12 | Qwen 3 30B A3B (Thinking) | 51.6 | #493 (Qwen 3 30B A3B) |
| 13 | DeepSeek V3 (0324) | 46.2 | #340 |
| 14 | Gemini 2.5 Flash | 42.85 | #245 |
| 15 | Qwen 3 235B A22B | 40.18 | #313 |
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-2025-05 · How It Works · Data refreshed daily, snapshot 2026-10-20.