Batching API calls in Custom GPT

Is there a way to batch API calls to an external service? For context, I’m currently building a Custom GPT that calls APIs of a Google Workspace account. Specifically, I’m calling to the Gmail API and Calendar API to help an account organize emails and create follow-up appointments.

Right now I’m doing a lot of excessive APIs calls. Once I get list of email IDs from a particular category, the Custom GPT then individually sends 1 get request for EACH email. I would like to reduce my API usage by instructing the system to batch multiple get requests.

Has anyone started experimenting with this?

I’m happy to provide a draft of my OpenAPI specification if it will help. Also , if you’re working on something similar, I’m happy to share other notes of my developments steps getting to this point.

For additional context: I’ve been able to iron out all OAuth 2.0 authentication issues between OpenAI and Google Workspace.

TL;DR: has anyone batched API calls from a Custom GPT to an external API service?

If the API you’re calling supports batch operations, you should be able to just define an action to that endpoint.

Can you share the endpoint you’re working with?