Any news about connecting a custom GPT to the API?

Hi everyone
Dudi,
I think I know what the developer had in mind and we are on the same topic - integrating gpt chat with API.
It was about the assistant to tell you how to make an assistant’s API from your gpt, these are just the terms, but for me the assistant could be llama2 or mistral. Check the API creator and gpt chat in APIopenai and explain exactly what you need, ask him for suggestions or if he would add anything, e.g. a nice atmosphere and willingness to understand will be rewarded when he understands you better.
He is very helpful and once he gets the outline, he will get an idea whether the task is going well :slight_smile: and the gpt chat is great, but there is no integration, but it has other advantages.
Personally, I will, I’m in the process of making a clone as a python API and it will probably work tomorrow and I’m here because I’m looking for additional solutions and the assistant gave me another option because I mentioned that I was closer to huggingface and showered me with more models. So believe me, I’m shocked at what an assistant can do after proper training. As I use it :slight_smile: :smiley: and I try to talk to a person and make jokes, after 15 minutes the assistant adjusted and I quote “perhaps consider a little less coffee next time to keep the chaos at bay! :smile:” without rushing and believe me, it’s even better I can do it and connect it to the API.
Kind regards as a new one here.

1 Like

+1. using customGPT via API will be very helpful for developer. assistant API is not the same at all. currently not support for example model 4 io.

This not happening ever, partly because there is zero reason for it but mostly because they are entirely different products.

2 Likes

I have trained a custom GPT and would like to have it as an API to integrate with my webapp. This will be a great feature, and allow developers to continue to pay for ChatGPT premium service. Assistant is of no use to us.

This has indeed been requested quite often, but OpenAI already offers the Assistants API for developers who want to integrate their solutions into an app. In fact there are many options available.

I think it’s important to understand that ChatGPT is intentionally designed as a closed ecosystem, and if connecting to a custom GPT via an external API were ever possible, it would very likely not be free or included in a basic subscription.

So yes, it is possible, but highly unlikely, given the existing options available to developers.

Hope this helps!

1 Like

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.