AWS adds usage-based metering for Amazon SageMaker AI bidirectional streaming

AWS announced usage-based metering for Amazon SageMaker AI bidirectional streaming inference, enabling model providers to report actual consumption during live sessions so AWS Marketplace software charges can track streaming usage instead of a single end-of-stream count. Deepgram is the launch partner, adding usage-based billing for speech-to-text and text-to-speech on AWS Marketplace. Amazon SageMaker AI endpoint infrastructure charges still apply separately while the endpoint is provisioned, and customers use the bidirectional streaming API to send audio through the HTTP/2 event stream.