solarized-18B-dpo: benchmark results

Provider: vicgalle. Access: Open.

Aggregate Rating 1432 ± 36, rank #4571 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - HellaSwag87.79Normalized accuracy (%) (10-shot)87.9
Open LLM Leaderboard v1 - TruthfulQA MC266.49MC2 (%) (0-shot)84.9
Open LLM Leaderboard v1 - ARC Challenge68.34Normalized accuracy (%) (25-shot)80.1
Open LLM Leaderboard v1 - WinoGrande80.51Accuracy (%) (5-shot)74.3
Open LLM Leaderboard v1 - MMLU63.89Accuracy (%) (5-shot)63.9
Open LLM Leaderboard v1 - GSM8K40.26Accuracy (%) (5-shot)55.5

Interactive version: theaggregate.ai/model?slug=solarized-18b-dpo · How It Works · Data refreshed daily, snapshot 2026-11-08.