phi-2-instruction [float16] [Adapter] [rev: 120e8a957f98]: benchmark results

Provider: Microsoft. Access: Open.

Aggregate Rating 1373 ± 34, rank #7069 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - GSM8K51.63Accuracy (%) (5-shot)63.6
Open LLM Leaderboard v1 - ARC Challenge61.09Normalized accuracy (%) (25-shot)51.2
Open LLM Leaderboard v1 - MMLU57.81Accuracy (%) (5-shot)41.9
Open LLM Leaderboard v1 - WinoGrande74.82Accuracy (%) (5-shot)38.9
Open LLM Leaderboard v1 - TruthfulQA MC245.1MC2 (%) (0-shot)29.7
Open LLM Leaderboard v1 - HellaSwag74.68Normalized accuracy (%) (10-shot)25.2

Interactive version: theaggregate.ai/model?slug=phi-2-instruction-float16-adapter-rev-120e8a957f98 · How It Works · Data refreshed daily, snapshot 2026-11-08.