PPO_Shygmalion-V8p4_Dev-6B: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1169 ± 15, rank #9773 of 10916 rated models, from 68 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MMLU-by-task - College Mathematics32Accuracy (%)53.7
MMLU-by-task - Moral Scenarios25.81Accuracy (%)50.5
MMLU-by-task - College Physics24.51Accuracy (%)48.1
MMLU-by-task - Econometrics28.07Accuracy (%)47.3
MMLU-by-task - High School Mathematics26.67Accuracy (%)45.9
MMLU-by-task - Security Studies41.22Accuracy (%)42
MMLU-by-task - Professional Law30.7Accuracy (%)39.9
MMLU-by-task - Abstract Algebra28Accuracy (%)39.6
MMLU-by-task - Astronomy35.53Accuracy (%)38.3
MMLU-by-task - College Medicine31.21Accuracy (%)35.2
MMLU-by-task - High School Macroeconomics35.13Accuracy (%)35
MMLU-by-task - International Law41.32Accuracy (%)34.7

Interactive version: theaggregate.ai/model?slug=ppo-shygmalion-v8p4-dev-6b · How It Works · Data refreshed daily, snapshot 2026-11-08.