Phi 4 Reasoning (verifiable): benchmark results

Provider: Other.

Aggregate Rating 1573 ± 27, rank #1360 of 10916 rated models, from 1 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Medmarks0.48Win Rate38.5

Interactive version: theaggregate.ai/model?slug=phi-4-reasoning-verifiable · How It Works · Data refreshed daily, snapshot 2026-11-08.