Cross region inference in Amazon Q Developer

Amazon Q Developer is powered by Amazon Bedrock, and uses cross-region inference to distribute traffic across different AWS Regions to enhance large language model (LLM) inference performance and reliability. With cross-region inference, you get:

  • Increased throughput and resilience during high demand periods

  • Improved performance

  • Access to newly launched Amazon Q Developer capabilities and features that rely on the most powerful LLMs hosted on Amazon Bedrock

Today, regardless of where you use Amazon Q Developer, your data is processed in a US Region unless you're using Amazon Q generative SQL outside the US regions. With cross-region inferencing, your requests to Amazon Q Developer may be processed in any of our US regions (currently US East (N. Virginia) Region, the US West (Oregon) Region, or the US East (Ohio) Region), even if you're using Amazon Q Developer in a different AWS Region. For information on where data is stored during processing, see Data protection. For information on where you can use Amazon Q Developer, see Supported Regions for Amazon Q Developer. There's no additional cost for using cross-region inference.

If you use Amazon Q generative SQL outside the US regions, your data may be processed locally or across Regions to support cross-region inference, depending on your location. If you choose the Asia Pacific (Seoul) AWS Region, cross-region inference would be used for inference requests due to model availability.