Phi-3.5-mini-instruct: benchmark results
Microsoft's 3.8B MIT-licensed small model with a 128K context and improved multilingual coverage. Provider: Microsoft. Released 2024-08-20. Access: Open.
Aggregate Rating 1416 ± 9, rank #1167 of 1656 rated models, from 300 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Stats (SQuALITY) | 24.3 | Score (%) | 100 |
| LIBRA - MatreshkaYesNo | 73.42 | Dataset Total Score (%) | 91.7 |
| Open Japanese LLM - Xlsum JA Bert Score JA F1 | 70.67 | Score (%) | 88.8 |
| LLM Stats (Social IQa) | 74.7 | Score (%) | 87.5 |
| GSM8K | 86.2 | Accuracy (%) | 87.1 |
| Open LLM Leaderboard - GPQA | 33.98 | Score | 84.4 |
| Enkrypt AI - Jailbreak Risk | 4.28 | Risk Score | 81.6 |
| Open Japanese LLM - Xlsum JA Rouge1 | 29.95 | Score (%) | 80.8 |
| Open Japanese LLM - Xlsum JA RougeLsum | 25.34 | Score (%) | 80.3 |
| BankMathBench (English) - Intermediate | 6.4 | Exact-match accuracy (%) on the 423 English intermediate (co | 77.3 |
| Open Japanese LLM - SUM | 10.73 | Score (%) | 76.6 |
| Open Japanese LLM - Xlsum JA Rouge2 | 10.73 | Score (%) | 76.6 |
Interactive version: theaggregate.ai/model?slug=phi-3-5-mini-instruct · How It Works · Data refreshed daily, snapshot 2026-10-20.