Brief
Amazon Bedrock prompt caching cuts repeated input costs by up to 90%
Amazon Bedrock's prompt caching can reduce input token costs by up to 90% when the same context is sent repeatedly to foundation models.
The feature targets a common pattern: sending the same long context—such as a system prompt or tool definition—over and over. By caching that context, you avoid paying full price for those tokens on each call. The source describes six practical scenarios using the Converse API: message content, system prompt, tool definition, mixed TTL, tenant isolation, and LangChain integration. The useful detail is the cost reduction for repeated contexts, which directly addresses the reader question of what this does that previous options could not.
Source details and supporting facts
Each line is stated by the page named above it.
Stated by aws.amazon.com
- Prompt caching in Amazon Bedrock can cut input token costs by up to 90% when you repeatedly send the same context to foundation models.
- The post walks through six practical prompt caching scenarios using the Converse API: message content, system prompt, tool definition, mixed TTL, tenant isolation, and LangChain integration.
Sources
- AWS Machine Learning BlogText stored 15 September 2026
How this story was checked. Written from the 1 page listed above, stored 15 September 2026; claims checked against that stored text on 15 September 2026.
What that means
- 2 of 2 reported statements were confirmed against the page that carries them; the rest were removed rather than published.
- Figures in the text were required to appear in the stored source text: yes. Identifiers: yes.
- The check reads stored text only: no claim rests on a fresh look that did not happen.
- Where the reporting was silent, the text says so instead of filling the gap.