phi-2-gpo-renew2-b0.001-v4-i1 [Adapter]: benchmark results

Provider: Microsoft. Access: Open.

Aggregate Rating 1374 ± 34, rank #7042 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - GSM8K55.5Accuracy (%) (5-shot)67.1
Open LLM Leaderboard v1 - ARC Challenge61.18Normalized accuracy (%) (25-shot)51.6
Open LLM Leaderboard v1 - MMLU58.11Accuracy (%) (5-shot)42.6
Open LLM Leaderboard v1 - WinoGrande73.95Accuracy (%) (5-shot)34.2
Open LLM Leaderboard v1 - TruthfulQA MC244.36MC2 (%) (0-shot)26.9
Open LLM Leaderboard v1 - HellaSwag74.92Normalized accuracy (%) (10-shot)25.5

Interactive version: theaggregate.ai/model?slug=phi-2-gpo-renew2-b0-001-v4-i1-adapter · How It Works · Data refreshed daily, snapshot 2026-11-08.