Llama 3.1 8B Ita: benchmark results
Michele Montebovi's (DeepMount00) Italian-language fine-tune of Llama 3.1 8B Instruct. Provider: DeepMount00. Released 2024-07-23. Access: Open.
Aggregate Rating 1496 ± 18, rank #797 of 1656 rated models, from 28 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - IFEval | 79.17 | Score | 96.4 |
| Open Ko-LLM Leaderboard | 48.31 | Average (%) | 95.9 |
| Open Korean LLM - Ko-IFEval | 53.34 | Strict accuracy, prompt/instruction mean (%) | 95.9 |
| Open Korean LLM Leaderboard | 48.31 | Average Score (%) | 95.9 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 60.88 | Exact match, flexible extract (%) | 94.5 |
| Pinocchio Italian - Cultura | 64.67 | Accuracy (%) | 88.6 |
| Pinocchio Italian - Diritto | 57.96 | Accuracy (%) | 88.6 |
| Pinocchio Italian - Generale | 61.27 | Accuracy (%) | 88.6 |
| Open Korean LLM - KorNAT-CKA | 31.86 | Normalized accuracy (%) | 87.8 |
| Open Korean LLM - Ko-EQ-Bench | 36.81 | EQ-Bench score | 87.3 |
| Pinocchio Italian Leaderboard | 58.5 | Average Accuracy (%) | 86.4 |
| Open LLM Leaderboard - MuSR | 44.87 | Score | 82.7 |
Interactive version: theaggregate.ai/model?slug=llama-3-1-8b-ita · How It Works · Data refreshed daily, snapshot 2026-10-20.