LLM2014 Logic 2026-01: leaderboard
LLM2014 monthly logic benchmark for 2026-01, reporting aggregate logical-reasoning performance on the source score scale.
Metric: Median Score. Source: raw.githubusercontent.com. Saturation forecast: Around January 2027. 45 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 (High) | 80.71 | #110 (GPT-5.2) |
| 2 | Gemini 3 Pro | 75.64 | #80 |
| 3 | Gemini 3 Flash (High) | 73.32 | #99 (Gemini 3 Flash) |
| 4 | GPT-5 Mini (High) | 71.43 | #186 (GPT-5 Mini) |
| 5 | Kimi K2.5 (Thinking) | 65.21 | #144 (Kimi K2.5) |
| 6 | Grok 4 | 63.91 | #181 |
| 7 | GLM-4.7 (Thinking) | 62.94 | #192 (GLM-4.7) |
| 8 | Claude Sonnet 4.5 (Thinking) | 60.83 | #142 (Claude Sonnet 4.5) |
| 9 | Kimi K2 (Thinking) | 55.69 | #243 (Kimi K2) |
| 10 | Grok 4.1 Fast (Reasoning) | 54.11 | #217 (Grok 4.1 Fast) |
| 11 | Qwen 3 Max (Preview) | 49.6 | #230 |
| 12 | Qwen 3 235B A22B 2507 (Thinking) | 46.71 | #260 (Qwen 3 235B A22B 2507) |
| 13 | ERNIE 5.0 | 41.43 | #233 |
| 14 | GPT-OSS-120B (High) | 38.39 | #341 (GPT-OSS-120B) |
| 15 | MiniMax-M2.1 | 36.72 | #289 |
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-2026-01 · How It Works · Data refreshed daily, snapshot 2026-10-20.