Custom GPT to Zendesk.

[meager echo]

Hello, we are trying to integrate custom data fed gpt model to zendesk, so that it could potentially answer that ticket. Since the new gpt feature is out, we are having some issues and there's little information on the internet.
I have some questions that i hope someone could help me here as im stuck.

1. In ZD what should we put as endpoint URL? Is it simply a https://api.openai.com/v1/chat/completions ?
2. Does the json file look ok?
   "data": {
   "ticket_id": "{{ticket.id}}",
   "subject": "{{ticket.subject}}",
   "requester": {
   "name": "{{ticket.requester.name}}",
   "email": "{{ticket.requester.email}}"
   },
   I guess we have to define a model to it right? But what to enter as "model": since we want to use our own specific made model? Any more tips is appriciated ❤️

[muted epoch]

Hi!
To create an assistant, you can make a request to https://api.openai.com/v1/assistants

To add a message to the thread or create a thread, you can make a request to https://api.openai.com/v1/threads

And to check whether a run has been completed or not you can make a request to https://api.openai.com/v1/threads/THREAD_ID/runs

You can find more information on https://platform.openai.com/docs/assistants/how-it-works/runs-and-run-steps

About the model, this is just the model that your assistant will be powered by, you can choose any GPT-3.5 or GPT-4 model, but I recommend using OpenAI's latest model gpt-4-1106-preview

For more information you can visit https://openai.com/pricing#language-models or https://platform.openai.com/docs/assistants/overview

I hope that answers your question, if you have any other questions you can always ask me!