AWS GovCloud (US) adds OpenAI GPT OSS and NVIDIA Nemotron models

Amazon Bedrock has expanded its model lineup in AWS GovCloud (US) to include OpenAI’s open-weight GPT OSS models (gpt-oss-120b and gpt-oss-20b) and NVIDIA’s Nemotron family (Nano 9B v2, Nano 12B v2, Nano 30B, and Super 120B). This launch addresses a critical need for U.S. government agencies and their contractors who require advanced AI capabilities without compromising security, compliance, or data residency obligations. AWS GovCloud (US) Regions are physically located in the U.S. and administered exclusively by U.S. citizens, meeting stringent frameworks like FedRAMP High, DoD SRG IL2/4/5, ITAR, and CJIS.

The new models enable the development of agentic applications and mission workflows, such as automated security control assessments, multi-document intelligence synthesis, and policy compliance checking, all within the AWS GovCloud (US) compliance boundary. NVIDIA Nemotron 3 Super 120B, a 120B-parameter hybrid mixture-of-experts (MoE) model, activates only 12B parameters per token, delivering up to 5x higher throughput than its predecessor and a 1M-token context window for long-term memory in multi-step tasks. Meanwhile, the Nemotron 3 Nano models (9B, 12B, and 30B parameters) offer smaller, compute-efficient options with up to 4x higher throughput. OpenAI’s GPT OSS models provide open-weight alternatives with strong performance for diverse generative AI use cases.