From the source
- AWS said on November 19, 2026 that PolyAI's Dialog-RSN-1 has joined the
Amazon SageMaker JumpStart catalog, giving customers one-click deployment, evaluation tools and four optimization profiles for the audio-native customer-service model.
- The endpoint runs inside the customer's own AWS account, and customers still subscribe to the model through AWS Marketplace.
- JumpStart creates a real-time inference endpoint on a preconfigured serving stack, removing manual container selection and the need to benchmark infrastructure combinations by hand, and the model card shows P50 latency, time to first token and throughput before deployment.
- The four profiles prioritize latency, throughput, cost or balanced general-purpose use, tying the serving configuration to call volume, responsiveness and spending.
- Dialog-RSN-1 processes caller audio directly and combines turn-taking, speech recognition, response generation and function calling, but it returns only streamed reply text or a function call and does not synthesize speech, so every deployment must add its own text-to-speech layer.
- Audio input is limited to base64-encoded PCM16 mono at 16 or 24 kHz, the context window is 32,000 tokens, the model is English-only, and connections have a 30-minute hard cap with a five-minute idle timeout.
- Real-time inference costs $45 per host hour on ml.p4d.24xlarge and $110 per host hour on ml.p5.48xlarge, and billing continues from when the endpoint enters service until deletion even when no client is connected.