phi_3: benchmark results
Provider: Other.
Aggregate Rating 1428 ± 36, rank #4744 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 67.88 | Accuracy (%) (5-shot) | 90 |
| Open LLM Leaderboard v1 - GSM8K | 69.6 | Accuracy (%) (5-shot) | 89.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 63.99 | Normalized accuracy (%) (25-shot) | 65.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 54.79 | MC2 (%) (0-shot) | 63.6 |
| Open LLM Leaderboard v1 - HellaSwag | 78.85 | Normalized accuracy (%) (10-shot) | 36 |
| Open LLM Leaderboard v1 - WinoGrande | 72.14 | Accuracy (%) (5-shot) | 28.3 |
Interactive version: theaggregate.ai/model?slug=phi-3 · How It Works · Data refreshed daily, snapshot 2026-11-08.