CZ-EVAL - Culture — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3 Opus | 92.45 |
| 2 | GPT-4 Preview (1106) | 90.23 |
| 3 | Mistral Large | 85.94 |
| 4 | Claude 3 Sonnet | 83.46 |
| 5 | Claude 3 Haiku | 81.64 |
| 6 | Gemini 1.0 Pro | 80.73 |
| 7 | Claude 2.1 | 79.82 |
| 8 | Mistral Medium | 77.34 |
| 9 | GPT-3.5 Turbo | 72.14 |
| 10 | Mixtral 8x7B Instruct | 71.22 |
| 11 | Command R | 70.44 |
| 12 | gemma-7B (IT) | 22.92 |
Interactive version: aibenchmarks.dev/benchmark?slug=cz-eval-culture · How the rankings work · Data refreshed daily, snapshot 2026-07-08.