EVALITA - text-entailment — leaderboard
Metric: CPS. Source: huggingface.co. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 85.07 |
| 2 | Qwen2.5-14B-Instruct-1M | 85.02 |
| 3 | calme-3.2-instruct-78B | 84.31 |
| 4 | Qwen 2.5 VL 32B Instruct | 83.7 |
| 5 | Qwen 2.5 7B Instruct | 83.65 |
| 6 | Mistral Small 3 | 82 |
| 7 | Gemma 3 27B (IT) | 81.13 |
| 8 | Mistral Large 2 (Nov) Instruct (2411) | 80.96 |
| 9 | Llama 3.3 70B Instruct | 80.79 |
| 10 | Gemma 2 9B (IT) | 79.97 |
| 11 | GLM-4.1V-9B (Thinking) | 79.79 |
| 12 | Gemma 3 12B (IT) | 79.52 |
| 13 | DeepSeek R1 Distill Llama 70B | 79.37 |
| 14 | Gemma 2 27B (IT) | 78.72 |
| 15 | Qwen 2.5 VL 7B Instruct | 78.23 |
Interactive version: theaggregate.ai/benchmark?slug=evalita-text-entailment · How the rankings work · Data refreshed daily, snapshot 2026-07-22.