LLM2014 Logic 2026-11: leaderboard
November 2026 LLM2014 Logic cohort, reported as the native median score. Source model, effort, refusal and estimated-result labels remain attached to their observations. Monthly editions stay separate. Source: https://raw.githubusercontent.com/llm2014/llm_benchmark/main/docs/data/logic/2026-11.csv
Metric: Median Score. Source: raw.githubusercontent.com. Saturation forecast: Around September 2027. 40 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra (xHigh) | 86.35 | #6 (GPT-6 Astra) |
| 2 | Gemini 4 Argon (High) | 69.13 | #11 (Gemini 4 Argon) |
| 3 | GPT-6.1 Sol (xHigh) | 67.79 | #10 (GPT-6.1 Sol) |
| 4 | Kimi K3.1 Pro (Max) | 59.87 | #28 (Kimi K3.1 Pro) |
| 5 | Muse Wonder (xHigh) | 56.06 | #27 (Muse Wonder) |
| 6 | DeepSeek V4.1 Flash (1017) (Max) | 55.14 | #38 (DeepSeek V4.1 Flash (1017)) |
| 7 | Claude Fable 5.1 (xHigh) | 55.12 | #14 (Claude Fable 5.1) |
| 8 | Claude Sonnet 5.5 (xHigh) | 53.83 | #18 (Claude Sonnet 5.5) |
| 9 | Grok 4.7 (xHigh) | 48.98 | #43 (Grok 4.7) |
| 10 | MiniMax-M3.1-Flash (Max) | 48.19 | #104 (MiniMax-M3.1-Flash) |
| 11 | Gemini 3.8 Flash (High) | 47.68 | #36 (Gemini 3.8 Flash) |
| 12 | GLM-5.3 (Max) | 44.15 | #62 (GLM-5.3) |
| 13 | DeepSeek V4 Pro (0813) (Max) | 43.86 | #81 (DeepSeek V4 Pro (0813)) |
| 14 | Muse Ember (xHigh) | 42.99 | #54 (Muse Ember) |
| 15 | Qwen 3.8 Max (xHigh) | 41.71 | #55 (Qwen 3.8 Max) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=llm2014-logic-2026-11 · How It Works · Data refreshed daily, snapshot 2026-11-12.