Replies: 1 comment
-
I think you can config the model before using. E.g. Bedrock sdk-python/src/strands/models/bedrock.py Line 58 in 34b1569 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Quick question about Prompt Caching when using bedrock. Looking here: https://strandsagents.com/latest/user-guide/concepts/model-providers/amazon-bedrock/#caching
Both system and tool caching is pretty obvious (and easy to use). I'm not 100% sure I understand messages caching though.
Right, now, I pretty simply use the agents like
response = myagent("Do some awesome agent stuff!")
. I think, if I read this right, I'd prime the agent with some of the normal bedrock json, and then set that to be cached?Would that mean that only the initial response is cached, or would cache's be created as the agent is doing its work?
It's entirely possible I'm not thinking of prompt caching quite right. Thanks for any help!
Beta Was this translation helpful? Give feedback.
All reactions