Phi-Delthanar: benchmark results
Provider: Microsoft.
Aggregate Rating 1287 ± 43, rank #8889 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 50.92 | MC2 (%) (0-shot) | 50 |
| Open LLM Leaderboard v1 - MMLU | 51.16 | Accuracy (%) (5-shot) | 30.5 |
| Open LLM Leaderboard v1 - ARC Challenge | 46.67 | Normalized accuracy (%) (25-shot) | 21.8 |
| Open LLM Leaderboard v1 - WinoGrande | 61.48 | Accuracy (%) (5-shot) | 14.6 |
| Open LLM Leaderboard v1 - HellaSwag | 60.19 | Normalized accuracy (%) (10-shot) | 14.3 |
| Open LLM Leaderboard v1 - GSM8K | 0 | Accuracy (%) (5-shot) | 4.5 |
Interactive version: theaggregate.ai/model?slug=phi-delthanar · How It Works · Data refreshed daily, snapshot 2026-11-08.