NVIDIA Nemotron 3 Nano now available on Amazon Bedrock serverless

Amazon Web Services has added NVIDIA Nemotron 3 Nano to Amazon Bedrock as a fully managed serverless model, making the compact language model available to developers without requiring any infrastructure provisioning. The addition means developers can call Nemotron 3 Nano through Bedrock's standard API on a pay-per-token basis, with AWS handling all underlying compute scaling automatically.

Previously, running NVIDIA's Nemotron models required self-managed infrastructure or direct access through NVIDIA's own APIs. The Bedrock serverless tier removes that barrier: developers pay only for tokens processed, with no minimum commitment and no idle compute costs. For teams already building on AWS, access requires no new service agreements -- Nemotron 3 Nano is available through the same Bedrock console and SDK used for other foundation models including Claude, Llama, and Mistral. The addition expands the Bedrock catalog's NVIDIA coverage and reflects the continued push by cloud providers to make specialized AI models accessible as managed, on-demand services rather than infrastructure challenges. Nemotron 3 Nano is designed for efficient inference at lower compute cost, making it a practical fit for the serverless, pay-as-you-go model that Bedrock offers.