LLM Stats (ZebraLogic): leaderboard
Logic puzzle benchmark based on zebra/Einstein puzzles requiring systematic constraint satisfaction and deductive reasoning.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 235B A22B (Thinking) | 97.3 | #238 (Qwen 3 VL 235B A22B) |
| 2 | LongCat-Flash-Thinking | 95.5 | #363 |
| 3 | Qwen 3 235B A22B 2507 Instruct | 95 | #300 |
| 4 | LongCat-Flash-Chat | 89.3 | #331 |
| 5 | Kimi K2 | 89 | #243 |
| 6 | Kimi K2 0905 | 89 | #288 |
| 7 | Qwen 3 14B | 88.5 | #537 |
| 8 | MiniMax M1 80k | 86.8 | #432 |
| 9 | MiniMax M1 40k | 80.1 | #683 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-zebralogic · How It Works · Data refreshed daily, snapshot 2026-10-20.