List your company

General Reasoning on GPQA D (Accuracy, Loss)

27.3AccuracyAdam
18.4620.75523.0525.345Apr 10, 2026

Evaluation results

MethodLinks
AdamModel scale=3B2026.0427.31.975
NexusModel scale=3B2026.0423.41.957
AdamModel scale=1B2026.0418.82.191
NexusModel scale=1B2026.0418.82.172