Systems fine-tuned from Nemotron 3 reached gold-medal level at both IOI 2026 and IMO 2026. The IOI system, Nemotron-3-Ultra-CC, scored 535.4 out of 600 under contest constraints, while the IMO system scored 30 of 42 points with proofs graded by official IMO graders. The teams used supervised fine-tuning, reinforcement learning, and feedback-driven inference, and released checkpoints, datasets, benchmarks, and inference pipelines on Hugging Face.
No score is assigned. Sources and their independence are shown in the citation chain below.