LLM2014 Logic 2026-05: leaderboard
LLM2014 monthly logic benchmark for 2026-05, reporting aggregate logical-reasoning performance on the source score scale.
Metric: Median Score. Source: raw.githubusercontent.com. Saturation forecast: Around January 2027. 43 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 (xHigh) | 80.47 | #31 (GPT-5.5) |
| 2 | Claude Opus 4.6 (High) | 76.48 | #63 (Claude Opus 4.6) |
| 3 | Gemini 3.5 Flash (High) | 73.93 | #60 (Gemini 3.5 Flash) |
| 4 | Claude Opus 4.8 (xHigh) | 68.32 | #37 (Claude Opus 4.8) |
| 5 | Gemini 3.1 Pro (Preview) (High) | 67.91 | #58 (Gemini 3.1 Pro (Preview)) |
| 6 | DeepSeek V4 Pro (Max) | 64.82 | #102 (DeepSeek V4 Pro) |
| 7 | Seed 2.0 Pro (High) | 57.4 | #118 (Seed 2.0 Pro) |
| 8 | GLM-5.1 | 55.59 | #97 |
| 9 | DeepSeek V4 Flash (Max) | 50.97 | #137 (DeepSeek V4 Flash) |
| 10 | Kimi K2.6 | 47.81 | #104 |
| 11 | Hy3-preview | 47.01 | #157 |
| 12 | Qwen 3.6 Plus | 44.99 | #120 |
| 13 | Claude Opus 4.6 | 43.29 | #63 |
| 14 | GPT-5.4 Mini (High) | 40.59 | #212 (GPT-5.4 Mini) |
| 15 | Claude Sonnet 4.5 | 39.93 | #142 |
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-2026-05 · How It Works · Data refreshed daily, snapshot 2026-10-20.