Replies: 2 comments
|
Thank you for your input!
So how much do we really save from this and is it worth the effort? I am not sure honestly, but also not strictly against it. Maybe we need to vote on that? |
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.
I'm already sort of doing this by using an LLM router which caches identical prompts and their outputs. But a way to reduce compute and cost overhead for this project would be to utilize cached results. Could even aggregate them on a server if needed, but local caching should be fine. This probably only applies directly to PKGBUILDs that don't change like devel versions based on git branch.
All reactions