Stheno-Mix-L2-20B: benchmark results

Provider: Other.

Aggregate Rating 1325 ± 15, rank #8248 of 10916 rated models, from 68 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MMLU-by-task - TruthfulQA MC251.8Accuracy (%)84.4
MMLU-by-task - TruthfulQA MC135.37Accuracy (%)82.6
MMLU-by-task - High School Chemistry44.33Accuracy (%)82.2
MMLU-by-task - Global Facts38Accuracy (%)81.7
MMLU-by-task - College Physics32.35Accuracy (%)80.6
MMLU-by-task - Electrical Engineering51.03Accuracy (%)79.8
MMLU-by-task - International Law73.55Accuracy (%)78.8
MMLU-by-task - Medical Genetics59Accuracy (%)75.4
MMLU-by-task - Computer Security68Accuracy (%)74.9
MMLU-by-task - Prehistory62.35Accuracy (%)72.5
MMLU-by-task - Security Studies61.63Accuracy (%)71.4
MMLU-by-task - College Computer Science45Accuracy (%)71.3

Interactive version: theaggregate.ai/model?slug=stheno-mix-l2-20b · How It Works · Data refreshed daily, snapshot 2026-11-08.