CZ-EVAL - Analytical — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 Preview (1106) | 51.59 |
| 2 | Claude 3 Opus | 47.26 |
| 3 | Mistral Large | 38.52 |
| 4 | Claude 2.1 | 38.04 |
| 5 | Claude 3 Sonnet | 37.85 |
| 6 | Claude 3 Haiku | 33.24 |
| 7 | Mistral Medium | 31.22 |
| 8 | GPT-3.5 Turbo | 30.45 |
| 9 | Gemini 1.0 Pro | 28.91 |
| 10 | Command R | 27.86 |
| 11 | Mixtral 8x7B Instruct | 25.26 |
| 12 | gemma-7B (IT) | 22.19 |
Interactive version: aibenchmarks.dev/benchmark?slug=cz-eval-analytical · How the rankings work · Data refreshed daily, snapshot 2026-07-08.