pygmalion-350m: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1126 ± 17, rank #10410 of 10916 rated models, from 68 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MMLU-by-task - High School Statistics47.22Accuracy (%)83.3
MMLU-by-task - Professional Medicine44.49Accuracy (%)42.8
MMLU-by-task - Conceptual Physics34.47Accuracy (%)37.8
MMLU-by-task - High School European History33.94Accuracy (%)36.1
MMLU-by-task - College Chemistry28Accuracy (%)35
MMLU-by-task - Security Studies37.14Accuracy (%)35
MMLU-by-task - Moral Disputes30.06Accuracy (%)32
MMLU-by-task - High School Geography34.34Accuracy (%)30.3
MMLU-by-task - Business Ethics31Accuracy (%)29.3
MMLU-by-task - TruthfulQA MC240.41Accuracy (%)28.8
MMLU-by-task - High School US History27.45Accuracy (%)28.3
MMLU-by-task - Sociology27.86Accuracy (%)28

Interactive version: theaggregate.ai/model?slug=pygmalion-350m · How It Works · Data refreshed daily, snapshot 2026-11-08.