近期结果表明，Nemotron 是构建世界级专家模型的强大且适应性强的基础。基于 Nemotron 3，我们的团队采用监督微调（SFT）、强化学习（RL）以及反馈驱动推理，构建了在 IMO 2026 和 IOI 2026 中均达到金牌水平的系统。IOI 的成绩源自一次实时、前瞻性的运行，其时间、互联网访问权限及提交约束条件与人类参赛者相同。该测试为非官方、无监督基准，未纳入 IOI 官方排名。

Our recent results show that Nemotron is a strong, adaptable foundation for building world-class specialist models. Starting from Nemotron 3, our teams used supervised fine-tuning (SFT), reinforcement learning (RL), and feedback-driven inference to create systems that reached gold-medal level at both IMO 2026 and IOI 2026. The IOI result came from a live, prospective run under the same time, internet-access, and submission constraints as human contestants. It was an unofficial, unsupervised benchmark and was not included in the official IOI ranking.

我们近期的研究结果表明，Nemotron 是构建世界级专家模型的强大且适应性强的基础。从 Nemotron 3 开始，我们的团队利用监督微调（SFT）、强化学习（RL）以及反馈驱动的推理技术，创建了能够在 IMO 2026 和 IOI 2026 中达到金牌水平的系统。IOI 的成绩源自一次实时的前瞻性运行，其时间限制、互联网访问权限及提交约束均与人类参赛者保持一致。该测试为非官方、无监督基准，未纳入 IOI 官方排名。IMO 系统提交的证明则由 IMO 官方评分员进行评判。“易于微调”不应仅意味着使检查点可训练，更应意味着一个强大的基础模型能够通过清晰、可复用的方案适配到高要求的领域。尽管训练和推理过程规模庞大，但其底层方法成熟且可复现。我们无需为每个挑战构建新的基础模型，而是对 Nemotron 进行了任务特定的专业化改造。 IMO 项目将这一理念应用于奥林匹克数学竞赛。基于 Nemotron 3 Ultra，我们分别使用 SFT 和 RL 训练了两个专家模型。在开发实验中，这两个后训练检查点的表现均优于通用可用模型。其中，SFT 检查点在首轮搜索中表现最强，而 RL 检查点则取得了最佳的单检查点整体结果。由于两者的优势互补，最终系统将这两个专家模型与通用模型结合使用。我们此前关于 IOI 2025 Hugging Face 的展示表明，测试时计算量可以将开源权重模型推向金牌级性能。新结果补充了一个重要环节：更好的专业化能为推理系统提供更优质的候选解、更有效的批评者以及更精准的优化。这一区别至关重要。这些奖牌并非仅靠微调获得，也并非仅靠暴力采样产生，而是源于模型、数据与推理循环的协同设计。 综上所述，IMO 和 IOI 为一个简单的理念提供了极具挑战性的证据：Nemotron 可以被微调为世界级的领域专家模型，随后通过透明的推理工作流进行组合，从而解决处于人类竞技前沿的问题。

Our recent results show that Nemotron is a strong, adaptable foundation for building world-class specialist models. Starting from Nemotron 3, our teams used supervised fine-tuning (SFT), reinforcement learning (RL), and feedback-driven inference to create systems that reached gold-medal level at both IMO 2026 and IOI 2026. The IOI result came from a live, prospective run under the same time, internet-access, and submission constraints as human contestants. It was an unofficial, unsupervised benchmark and was not included in the official IOI ranking. The IMO system’s submitted proofs were graded by official IMO graders. "Easy to fine-tune" should mean more than making a checkpoint trainable. It should mean that a capable foundation model can be adapted to a demanding domain with a clear, reusable recipe. The training and inference runs were substantial, but the underlying approach is familiar and reproducible. We did not need to build a new foundation model for every challenge. We specialized Nemotron for the task. The IMO project applied the same idea to olympiad mathematics. Starting from Nemotron 3 Ultra, we trained one specialist with SFT and another with RL. Both post-trained checkpoints outperformed the general-availability model in the development experiments. The SFT checkpoint was strongest in the first search round, while the RL checkpoint achieved the best overall single-checkpoint result. Their strengths were complementary, so the final system used both specialists alongside the general model. Our earlier IOI 2025 Hugging Face post showed how test-time compute can push open-weight models to gold-level performance. The new results add an important piece: better specialization gives the inference system better candidates, better critics, and better refinements. This distinction matters. The medals were not produced by fine-tuning alone, and they were not produced by brute-force sampling alone. They came from co-designing the model, the data, and the inference loop. Together, IMO and IOI provide unusually demanding evidence for a simple idea: Nemotron can be fine-tuned into world-class domain specialists, then composed with transparent inference workflows to solve problems at the frontier of human competition.