Marx-3B-V2: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1156 ± 16, rank #9874 of 10916 rated models, from 68 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MMLU-by-task - College Physics28.43Accuracy (%)67.4
MMLU-by-task - College Mathematics33Accuracy (%)61.7
MMLU-by-task - Moral Scenarios26.48Accuracy (%)54.4
MMLU-by-task - High School Physics29.14Accuracy (%)42.7
MMLU-by-task - High School Statistics35.19Accuracy (%)42.2
MMLU-by-task - HellaSwag54.85Accuracy (%)39.7
MMLU-by-task - Astronomy35.53Accuracy (%)38.3
MMLU-by-task - Anatomy37.04Accuracy (%)37.1
MMLU-by-task - College Computer Science33Accuracy (%)36.5
MMLU-by-task - High School European History32.73Accuracy (%)35.6
MMLU-by-task - ARC Challenge39.93Accuracy (%)34.1
MMLU-by-task - High School Mathematics25.93Accuracy (%)34.1

Interactive version: theaggregate.ai/model?slug=marx-3b-v2 · How It Works · Data refreshed daily, snapshot 2026-11-08.