LLM2014 Logic 2026-04: leaderboard
LLM2014 monthly logic benchmark for 2026-04, reporting aggregate logical-reasoning performance on the source score scale.
Metric: Median Score. Source: raw.githubusercontent.com. Saturation forecast: Around January 2027. 42 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 (xHigh) | 83.96 | #31 (GPT-5.5) |
| 2 | GPT-5.4 (High) | 79.56 | #79 (GPT-5.4) |
| 3 | Claude Opus 4.6 (Thinking) | 77.23 | #63 (Claude Opus 4.6) |
| 4 | Gemini 3.1 Pro (Preview) (High) | 72.16 | #58 (Gemini 3.1 Pro (Preview)) |
| 5 | DeepSeek V4 Pro (Max) | 70.49 | #102 (DeepSeek V4 Pro) |
| 6 | Seed 2.0 Pro (High) | 66.02 | #118 (Seed 2.0 Pro) |
| 7 | GLM-5.1 (Thinking) | 64.37 | #97 (GLM-5.1) |
| 8 | Qwen 3.6 Max (Preview) (Thinking) | 60.68 | |
| 9 | Gemini 3 Flash (High) | 58.87 | #99 (Gemini 3 Flash) |
| 10 | Kimi K2.6 (Thinking) | 57.33 | #104 (Kimi K2.6) |
| 11 | Hy3-preview | 56.63 | #157 |
| 12 | DeepSeek V4 Flash (Max) | 56.54 | #137 (DeepSeek V4 Flash) |
| 13 | Qwen 3.6 Plus (Thinking) | 55.35 | #120 (Qwen 3.6 Plus) |
| 14 | GPT-5.4 Mini (High) | 52.61 | #212 (GPT-5.4 Mini) |
| 15 | Claude Opus 4.6 | 51.58 | #63 |
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-2026-04 · How It Works · Data refreshed daily, snapshot 2026-10-20.