Amazon Bedrock Introduces Cross-Region Inference for OpenAI GPT-5.6 Models
Amazon Bedrock has launched cross-Region inference for three OpenAI GPT-5.6 model variants, enabling users to route requests across multiple AWS Regions to improve throughput and capacity management.
Original source published: August 20, 2026
Amazon Bedrock now supports cross-Region inference (CRIS) for three OpenAI GPT-5.6 model variants: Sol, Terra, and Luna. This feature allows users to route inference requests to a broader pool of compute resources beyond a single AWS Region, which can help maintain consistent performance and improve throughput under load. Users can utilize geographic inference profiles to keep data processing within specific boundaries or global inference profiles to access capacity across all supported commercial AWS Regions where the models are deployed.
These GPT-5.6 models support text and image inputs, a 1 million token context window, reasoning mode, and prompt caching. Developers can access these models via the OpenAI Responses API, the OpenAI Chat Completions API, or the Amazon Bedrock Converse API. This update is relevant for developers looking to scale AI workloads while managing capacity constraints or adhering to specific data residency requirements.