EVALITA - faq — leaderboard
Metric: CPS. Source: huggingface.co. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | LLaMAntino-3-ANITA-8B-Inst-DPO-ITA | 56.55 |
| 2 | Llama-3.1-SuperNova-Lite | 54.89 |
| 3 | granite-3.1-8B-instruct | 54.7 |
| 4 | Mistral Large 2 (Nov) Instruct (2411) | 54.69 |
| 5 | Mistral Small 3 | 54.55 |
| 6 | Qwen2.5-14B-Instruct-1M | 54.41 |
| 7 | aya-expanse-8B | 54.39 |
| 8 | Llama 3.3 70B Instruct | 54.3 |
| 9 | Phi-4 | 54.3 |
| 10 | Qwen 2.5 72B Instruct | 54 |
| 11 | Llama 4 Scout Instruct | 53.97 |
| 12 | InternVL3.5-8B | 53.86 |
| 13 | Qwen 2.5 VL 32B Instruct | 53.84 |
| 14 | Pixtral-12B | 53.83 |
| 15 | Llama 3.1 8B Instruct | 53.73 |
Interactive version: theaggregate.ai/benchmark?slug=evalita-faq · How the rankings work · Data refreshed daily, snapshot 2026-07-22.