Introducing usage-based pricing for Amazon SageMaker AI bidirectional streaming
AWS Marketplace now supports usage-based metering for Amazon SageMaker bidirectional streaming inference, so model providers can charge for what customers actually consume during a live stream. Deepgram is the launch partner, bringing usage-based speech-to-text and text-to-speech billing to AWS Marketplace.
First seen by Oriel on .
AI summary
Amazon SageMaker AI now supports usage-based AWS Marketplace software metering for bidirectional streaming inference, aligning charges with reported consumption while retaining separate endpoint infrastructure fees.
- Deepgram is the launch partner, offering usage-based Speech-to-Text and Text-to-Speech billing for real-time streaming workloads.
- SageMaker AI opens a companion metadata WebSocket alongside the inference stream, correlating both connections through X-Amzn-SageMaker-Request-Id.
- Providers send incremental JSON records using inference.count and ConsumedUnits, with ClientToken providing per-connection idempotency.
- Metadata frames are limited to 512 bytes, one record per frame, text only, and one message per second per connection.
- Providers can terminate inference when metering fails or preserve availability in data-only mode; malformed records may still be dropped and require billing reconciliation.
#amazon sagemaker ai #aws marketplace #bidirectional streaming #usage-based pricing
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| AWS Machine Learning Blog | 09f20b599e1ea4e477b4979d15326bbd992cf96d | https://aws.amazon.com/blogs/machine-learning/introducing-usage-based-pricing-for-amazon-sagemaker-ai-bidirectional-streaming/ Read discussion |