Amazon SageMaker AI now offers serverless model customization for NVIDIA Nemotron 3 models, starting with Nemotron 3 Nano (30B total parameters, 3B active) and Nemotron 3 Super (120B total parameters, 12B active). The Nemotron 3 family uses a hybrid Mamba-Transformer Mixture-of-Experts (MoE) architecture with native support for up to 1M-token context lengths, activating only a fraction of total parameters per forward pass to deliver high throughput and strong accuracy at lower compute cost. Available fine-tuning techniques include supervised fine-tuning (SFT), reinforcement learning with verifiable rewards (RLVR), and reinforcement learning with AI feedback (RLAIF), allowing adaptation to specific domains and workflows without provisioning or managing infrastructure. The models use multi-environment reinforcement learning through NeMo Gym for alignment to real-world, multi-step agentic tasks across domains such as coding, reasoning, and long-context analysis.