Phi 1 5 — Benchmarks
Benchmark scores for Phi 1 5 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
Knowledge
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| HellaSwag | 47.6% | #40 / 42 | #70 / 76 |
| ARC Challenge | 44.4% | #36 / 51 | #54 / 77 |
| BoolQ | 75.8% | #20 / 33 | #50 / 77 |
| OpenBookQA | 37.2% | #18 / 21 | #36 / 42 |
| MMLU | 37.6% | #68 / 77 | #122 / 136 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| WinoGrande | 73.4% | #18 / 46 | #40 / 80 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.