LLaMAntino-2-7B-hf-ITA: benchmark results
Provider: Meta. Released 2023-12-14. Access: Open.
Unified ELO 1307 ± 1, rank #1340 of 1392 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EVALITA - summarization-fanpage | 26.7 | CPS | 51.1 |
| EVALITA - evalita NER | 29.28 | CPS | 29.8 |
| EVALITA - lexical-substitution | 19.74 | CPS | 25.5 |
| Open Italian LLM - MMLU-Pro (IT) | 15.25 | Accuracy (%) | 14.3 |
| EVALITA - relation-extraction | 28.86 | CPS | 12.8 |
| EVALITA - word-in-context | 55.94 | CPS | 12.8 |
| Pinocchio Italian - Diritto | 32.11 | Accuracy (%) | 9.1 |
| EVALITA - hate-speech-detection | 57 | CPS | 8.5 |
| EVALITA LLM Leaderboard | 37.86 | Average CPS | 8.5 |
| Pinocchio Italian - Cultura | 29.62 | Accuracy (%) | 6.8 |
| Pinocchio Italian - Generale | 31.6 | Accuracy (%) | 6.8 |
| Pinocchio Italian - Lingua Straniera | 31.37 | Accuracy (%) | 6.8 |
Interactive version: theaggregate.ai/model?slug=llamantino-2-7b-hf-ita · How It Works · Data refreshed daily, snapshot 2026-09-05.