CZ-EVAL - Verbal — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3 Opus | 80.18 |
| 2 | GPT-4 Preview (1106) | 72.67 |
| 3 | Claude 2.1 | 63.36 |
| 4 | Mistral Large | 62.76 |
| 5 | Claude 3 Haiku | 62.46 |
| 6 | Claude 3 Sonnet | 61.26 |
| 7 | Gemini 1.0 Pro | 60.96 |
| 8 | Mistral Medium | 50.45 |
| 9 | GPT-3.5 Turbo | 44.74 |
| 10 | Command R | 44.44 |
| 11 | Mixtral 8x7B Instruct | 43.54 |
| 12 | gemma-7B (IT) | 22.82 |
Interactive version: aibenchmarks.dev/benchmark?slug=cz-eval-verbal · How the rankings work · Data refreshed daily, snapshot 2026-07-08.