LLM2014 Logic 2025-10: leaderboard
LLM2014 monthly logic benchmark for 2025-10, reporting aggregate logical-reasoning performance on the source score scale.
Metric: Median Score. Source: raw.githubusercontent.com. Saturation forecast: Around January 2027. 50 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 (High) | 86.25 | #98 (GPT-5) |
| 2 | GPT-5 Mini (High) | 79.37 | #186 (GPT-5 Mini) |
| 3 | Grok 4 | 74.97 | #181 |
| 4 | Claude Sonnet 4.5 (Thinking) | 69.78 | #142 (Claude Sonnet 4.5) |
| 5 | Grok 4 Fast (Reasoning) | 67.76 | #254 (Grok 4 Fast) |
| 6 | Qwen 3 235B A22B 2507 (Thinking) | 54.18 | #260 (Qwen 3 235B A22B 2507) |
| 7 | DeepSeek V3.2 (Thinking) | 54.16 | #205 (DeepSeek V3.2) |
| 8 | GPT-OSS-120B (High) | 53.24 | #341 (GPT-OSS-120B) |
| 9 | Qwen 3 Max | 52.61 | #208 |
| 10 | Gemini 2.5 Pro | 52.54 | #152 |
| 11 | Seed-1.6 (Thinking) | 50.35 | #265 (Seed-1.6) |
| 12 | Qwen 3 30B A3B 2507 (Thinking) | 47.41 | #376 (Qwen 3 30B A3B 2507) |
| 13 | Ring-1T | 46.07 | #319 |
| 14 | Qwen 3 235B A22B 2507 | 43.97 | #260 |
| 15 | Qwen 3 Next 80B A3B (Thinking) | 43.77 | #320 (Qwen 3 Next 80B A3B) |
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-2025-10 · How It Works · Data refreshed daily, snapshot 2026-10-20.