CZ-EVAL - Critical Thinking — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3 Opus | 76.45 |
| 2 | GPT-4 Preview (1106) | 70.65 |
| 3 | Claude 3 Sonnet | 68.89 |
| 4 | Mistral Large | 66.78 |
| 5 | Claude 2.1 | 64.5 |
| 6 | Gemini 1.0 Pro | 60.98 |
| 7 | Claude 3 Haiku | 60.46 |
| 8 | Mistral Medium | 59.58 |
| 9 | Command R | 53.43 |
| 10 | Mixtral 8x7B Instruct | 51.14 |
| 11 | GPT-3.5 Turbo | 49.91 |
| 12 | gemma-7B (IT) | 27.94 |
Interactive version: aibenchmarks.dev/benchmark?slug=cz-eval-critical-thinking · How the rankings work · Data refreshed daily, snapshot 2026-07-08.