zephyr-gemma-rpo: benchmark results

Provider: HuggingFace. Access: Open.

Aggregate Rating 1375 ± 33, rank #6991 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - HellaSwag83.5Normalized accuracy (%) (10-shot)59.2
Open LLM Leaderboard v1 - GSM8K42.23Accuracy (%) (5-shot)56.9
Open LLM Leaderboard v1 - TruthfulQA MC249.72MC2 (%) (0-shot)45.8
Open LLM Leaderboard v1 - MMLU59.27Accuracy (%) (5-shot)44.8
Open LLM Leaderboard v1 - ARC Challenge56.91Normalized accuracy (%) (25-shot)36.3
Open LLM Leaderboard v1 - WinoGrande73.95Accuracy (%) (5-shot)34.2

Interactive version: theaggregate.ai/model?slug=zephyr-gemma-rpo · How It Works · Data refreshed daily, snapshot 2026-11-08.