PolyAI Dialog-RSN-1 Model Now Available via SageMaker JumpStart and AWS Marketplace
PolyAI Dialog-RSN-1, a real-time reasoning model for complex dialog, is now directly deployable in AWS customer environments through Amazon SageMaker JumpStart and AWS Marketplace. This release enables secure, one-click deployment and new integration options for conversational AI workloads.
What changed?
PolyAI Dialog-RSN-1 is now available for deployment on Amazon SageMaker JumpStart and AWS Marketplace. It can be launched as a real-time inference endpoint within an AWS account, using pre-configured deployment workflows that require no manual container management or benchmarking. The model accepts direct audio input, reasons over caller needs, and responds in real time, all within a single model invocation. It supports several optimization profiles for latency, throughput, and cost, and fully supports data residency by confining processing and storage within the customer's AWS account. Dialog-RSN-1 also provides integration via the OpenAI Realtime protocol through a PolyAI bridge.

Why does it matter to an everyday developer?
This update simplifies the deployment and evaluation of advanced dialog AI in AWS environments. Developers can now adopt PolyAI's reasoning-centric voice model without setting up complex infrastructure or managing custom containers. The direct audio input and real-time response eliminate the need for separate speech-to-text or dialog orchestration services, reducing both latency and failure points within customer-facing workflows. Data residency and auditability are straightforward since all data remains inside the AWS account boundary, aligned with enterprise compliance requirements. One-click deployment and sample notebooks lower the barrier to trial and integration for teams building conversational AI solutions.
What can the developer do now?
- Discover the PolyAI Dialog-RSN-1 model in SageMaker JumpStart or on AWS Marketplace.
- Deploy the model as a real-time endpoint in your AWS account with one click from the SageMaker AI console or using the SageMaker Python SDK.

