- Who
- Amazon Web Services
- What
- announced usage-based metering for Amazon SageMaker AI bidirectional streaming inference
- Quote
Today, we’re announcing usage-based metering for Amazon SageMaker bidirectional (bidi) streaming inference, so model providers can report actual usage for the AWS Marketplace software charge.
What happened
AWS announced usage-based metering for Amazon SageMaker AI bidirectional streaming inference, allowing model providers to report consumed units for AWS Marketplace software charges during live sessions. The integration uses a companion WebSocket for metering records, while endpoint infrastructure charges remain separate. Deepgram is the launch partner for usage-based Speech-to-Text and Text-to-Speech billing.
Read next
See the full storyAWS Adds Usage-Based Metering for SageMaker AI Bidirectional StreamingThe story so far
AWS added usage-based metering for Amazon SageMaker AI bidirectional streaming inference, letting model providers report usage for AWS Marketplace software charges during live sessions. Deepgram is the launch partner, offering usage-based billing for Speech-to-Text and Text-to-Speech. SageMaker endpoint infrastructure charges remain separate. Providers send metering records over a companion WebSocket metadata stream, limited to one message per second per connection and 512 bytes per message. If the stream fails to connect or drops during a session, inference continues by default.