LLM2014 Logic 2026-03: leaderboard
Metric: Median Score. Source: raw.githubusercontent.com. 42 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 (High) | 78.85 |
| 2 | Claude Opus 4.6 (Thinking) | 77.23 |
| 3 | Gemini 3.1 Pro (Preview) (High) | 75.02 |
| 4 | Seed 2.0 Pro (High) | 68.36 |
| 5 | Gemini 3 Flash (High) | 65.29 |
| 6 | Kimi K2.5 (Thinking) | 59.56 |
| 7 | GLM-5 (Thinking) | 55.08 |
| 8 | GPT-5.4 Mini (High) | 54.45 |
| 9 | Claude Opus 4.6 | 53.93 |
| 10 | Claude Sonnet 4.5 (Thinking) | 52.81 |
| 11 | Qwen 3.5 Plus (Thinking) | 51.81 |
| 12 | MiniMax-M2.7 | 49.52 |
| 13 | Step 3.5 Flash | 39.99 |
| 14 | Qwen 3.5 Plus | 39.29 |
| 15 | LongCat-Flash-Thinking-2601 | 37.69 |
Interactive version: theaggregate.ai/benchmark?slug=llm2014-logic-2026-03 · How It Works · Data refreshed daily, snapshot 2026-09-05.