LLM2014 Logic 2026-04: leaderboard
Metric: Median Score. Source: raw.githubusercontent.com. 42 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 (xHigh) | 83.96 |
| 2 | GPT-5.4 (High) | 79.56 |
| 3 | Claude Opus 4.6 (Thinking) | 77.23 |
| 4 | Gemini 3.1 Pro (Preview) (High) | 72.16 |
| 5 | DeepSeek V4 Pro (Max) | 70.49 |
| 6 | Seed 2.0 Pro (High) | 66.02 |
| 7 | GLM-5.1 (Thinking) | 64.37 |
| 8 | Gemini 3 Flash (High) | 58.87 |
| 9 | Kimi K2.6 (Thinking) | 57.33 |
| 10 | Hy3-preview | 56.63 |
| 11 | DeepSeek V4 Flash (Max) | 56.54 |
| 12 | GPT-5.4 Mini (High) | 52.61 |
| 13 | Claude Opus 4.6 | 51.58 |
| 14 | MiniMax-M2.7 | 47.37 |
| 15 | Claude Sonnet 4.5 (Thinking) | 46.89 |
Interactive version: theaggregate.ai/benchmark?slug=llm2014-logic-2026-04 · How It Works · Data refreshed daily, snapshot 2026-09-05.