List your company

Robust Reasoning on MMLU-Pro (Accuracy)

21AccuracyNITP
6.585610.327814.0717.8122May 24, 2026

Evaluation results

MethodLinks
NITPModel scale=9bA1b, Evaluation=few-shot, Context length=81922026.0521
NTPModel scale=9bA1b, Evaluation=few-shot, Context length=81922026.0515.29
NITPModel scale=3bA0.5b, Evaluation=few-shot, Context length=81922026.0512.29
NTPModel scale=3bA0.5b, Evaluation=few-shot, Context length=81922026.0511
NITPModel scale=1.9bA0.3b, Evaluation=few-shot, Context length=81922026.057.47
NTPModel scale=1.9bA0.3b, Evaluation=few-shot, Context length=81922026.057.14