Amazon Bedrock open-weight models gain explicit prompt caching
Start here
Amazon Bedrock's open-weight models now support explicit prompt caching, in addition to the automatic implicit caching already available, according to an AWS Machine Learning Blog post. Applications can reuse a matching prompt prefix across requests instead of reprocessing it, and the post walks through both modes via the Chat Completions API and shows implicit caching in an OpenCode coding-agent session. Costs depend on the operation: cache reads are billed at a lower rate than standard input, so savings grow each time a prefix is reused, while cache writes may be billed at a higher rate than standard input depending on the model. Model availability and prompt-caching support also vary by AWS Region, so users are told to check the model card for supported Regions.
Generated from reports · updated 25 min ago
Report timeline
Follow the reports to see the event from different sides.
- AWS Machine Learning BlogUsing prompt caching with open-weight models on Amazon Bedrock
Amazon Bedrock's open-weight models now support explicit prompt caching alongside automatic implicit caching, letting applications reuse a matching prompt prefix across requests instead of reprocessing it. Cache reads are billed at a lower rate than standard input, and the post demonstrates both modes through the Chat Completions API plus implicit caching in an OpenCode coding agent session.
Heat for this event
Not enough continuous observations yet to draw a trend.