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

BenchmarkScoreOpen rankAll models
HellaSwag47.6%#40 / 42#70 / 76
ARC Challenge44.4%#36 / 51#54 / 77
BoolQ75.8%#20 / 33#50 / 77
OpenBookQA37.2%#18 / 21#36 / 42
MMLU37.6%#68 / 77#122 / 136

Reasoning

BenchmarkScoreOpen rankAll models
WinoGrande73.4%#18 / 46#40 / 80

Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.

Phi 1 5 Benchmarks — Scores & Rankings | llmrun