phi-gemma-nlaf-v1 [Adapter]: benchmark results
Provider: Microsoft. Access: Open.
Aggregate Rating 1230 ± 33, rank #9415 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 45.86 | MC2 (%) (0-shot) | 32.6 |
| Open LLM Leaderboard v1 - GSM8K | 5.84 | Accuracy (%) (5-shot) | 24.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 43.86 | Normalized accuracy (%) (25-shot) | 19.6 |
| Open LLM Leaderboard v1 - MMLU | 37.58 | Accuracy (%) (5-shot) | 19.3 |
| Open LLM Leaderboard v1 - HellaSwag | 62.7 | Normalized accuracy (%) (10-shot) | 16.1 |
| Open LLM Leaderboard v1 - WinoGrande | 61.17 | Accuracy (%) (5-shot) | 14.3 |
Interactive version: theaggregate.ai/model?slug=phi-gemma-nlaf-v1-adapter · How It Works · Data refreshed daily, snapshot 2026-11-08.