One Model Family, Two Gold-Level Results: Fine-Tuning Nemotron for IOI and IMO

Nemotron models have achieved gold-level results in both the International Olympiad in Informatics (IOI) and the International Mathematical Olympiad (IMO). For IOI 2026, Nemotron-3-Ultra-CC, fine-tuned with SFT and GenCorrect, scored 535.4/600, surpassing the gold threshold of 361.12 and the top human score of 498.27. In IMO 2026, Nemotron 3 Ultra, utilizing SFT and RL checkpoints in a generate-verify-refine system, scored 30/42, exceeding the official gold threshold of 29.

This report uniquely highlights Nemotron's gold-level performance in both IOI and IMO, unlike previous models that typically excel in only one of these distinct competitions.