LiveBench Zebra Puzzle — leaderboard
Metric: Score. Source: livebench.ai. 73 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20240620) | 48 |
| 2 | GPT-4o ChatGPT | 46 |
| 3 | Command-R+ | 46 |
| 4 | GPT-4 Preview (0125) | 44 |
| 5 | Gemma 2 27B (IT) | 42 |
| 6 | GPT-4o (2024-08-06) | 42 |
| 7 | GPT-4 (0613) | 42 |
| 8 | DeepSeek Coder V2 | 42 |
| 9 | Qwen 2 72B Instruct | 40 |
| 10 | GPT-4o (2024-05-13) | 40 |
| 11 | Mistral Large | 40 |
| 12 | Phi-3-small-8k-instruct | 40 |
| 13 | GPT-4 Turbo | 38 |
| 14 | Command-R+ (08-2024) | 38 |
| 15 | Command R | 38 |
Interactive version: theaggregate.ai/benchmark?slug=livebench-zebra-puzzle · How the rankings work · Data refreshed daily, snapshot 2026-07-22.