Issue 2026-10-07 · Industry · 研究 · 模型发布
Fine-Tuned Nemotron Reaches Gold Level at IOI and IMO

Starting from Nemotron 3, teams used supervised fine-tuning, reinforcement learning and an inference loop to build specialists that hit gold-medal level at both IMO 2026 and IOI 2026. The IOI run was an unofficial live benchmark under human-equivalent time and submission limits, while IMO proofs were graded by official IMO graders.
Read original ↗