Pinocchio Italian Leaderboard: leaderboard
Italian-language evaluation covering general knowledge, logic, foreign language, math and science, law, and culture categories.
Metric: Average Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 45 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 2 27B (IT) | 70.97 | #645 |
| 2 | Lexora-Medium-7B | 68.98 | #773 |
| 3 | Gemma 2 9B (IT) | 68.37 | #788 |
| 4 | Phi-3-medium-4k-instruct | 65.57 | #952 |
| 5 | Llama 3.1 8B Ita | 58.5 | #797 |
| 6 | Llama 3.1 8B Instruct | 56.93 | #1035 |
| 7 | Llama 3 8B Ita | 56.37 | #926 |
| 8 | LLaMAntino-3-ANITA-8B-Inst-DPO-ITA | 55.38 | #1009 |
| 9 | Llama 3 8B Instruct | 55.17 | #1129 |
| 10 | Volare | 55.14 | #1213 |
| 11 | Qwen 2 1.5B Instruct | 51.44 | #1426 |
| 12 | Gemma 2 2B (IT) | 51.21 | #1341 |
| 13 | Llama 3 8B 4bit UltraChat Ita | 47.02 | #1280 |
| 14 | Qwen 2 1.5B | 46.62 | #1433 |
| 15 | Mistral-7B-v0.1 | 44.14 | #1326 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=pinocchio-italian-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.