Would be great to ping an LLM or AI Platform API with a query and let us know if it's down - i know we can build this, but you have this mostly built already www.latimer.ai - we have 14 models as RAG and often the foundation model will be deprecated without notice - thx
1 Comment
Thanks for this suggestion John, but this seems way too specific of a need.
I can't implement and maintain a specific check type for every LLM API out there.
You use and know their API so you can monitor the endpoint of your choice for those LLM using your updown account if you want.
Sign in to comment