Post-Training Language Models for Gold-Medal Performance in Coding Competitions

The coding companion to NVIDIA's IMO gold work: an end-to-end specialisation pipeline of problem curation, synthetic reasoning traces, SFT, and RL over 22,000 curated problems that produces Nemotron-3-Nano-CC (30B-A3B, SFT plus RL) and Nemotron-3-Ultra-CC (550B-A55B, SFT only), plus GenCorrect, a feedback-driven test-time strategy that generates, evaluates, and refines diverse solutions. On IOI 2025 Nano-CC rises from 130 to 291 points after the pipeline, and on a prospective IOI 2026 evaluation the system reaches 535.4 of 600 against a gold threshold of 361. The Ultra-CC checkpoint is described as to be released; no repositories had appeared by 17 September 2026.