pythia-1B-dpo: benchmark results

Provider: EleutherAI.

Aggregate Rating 1123 ± 33, rank #10469 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - GSM8K1.67Accuracy (%) (5-shot)17.9
Open LLM Leaderboard v1 - HellaSwag49.24Normalized accuracy (%) (10-shot)10
Open LLM Leaderboard v1 - ARC Challenge30.12Normalized accuracy (%) (25-shot)9.1
Open LLM Leaderboard v1 - WinoGrande54.06Accuracy (%) (5-shot)8.2
Open LLM Leaderboard v1 - TruthfulQA MC237.2MC2 (%) (0-shot)5.2
Open LLM Leaderboard v1 - MMLU24.24Accuracy (%) (5-shot)3.3

Interactive version: theaggregate.ai/model?slug=pythia-1b-dpo · How It Works · Data refreshed daily, snapshot 2026-11-08.