I’m considering building a custom GPT model using OpenAI’s GPT builder

Community
,
1

I’d like to know if, once I create this custom model, I will be able to communicate with it using the OpenAI API.

Specifically, I want to:

  1. Use the API to send prompts and receive responses from my custom GPT model.
  2. Integrate this custom model into my application, just like I would with the standard OpenAI GPT models.

Could anyone provide insight into how this process works? Are there any special steps or configurations needed to enable API access for a custom GPT model?

2

Hi there @shahrukhkarlal and welcome to the Dev Community!

CustomGPTs can only be accessed and used via the ChatGPT interface. It is not possible to access them via API or to integrate them directly into your own application. You should look into OpenAI’s Assistants. They are designed for just this purpose.

Here is a good overview contrasting the two options:

https://help.openai.com/en/articles/8673914-gpts-vs-assistants

Additionally, here is the link to the official guide on Assistants:

https://platform.openai.com/docs/assistants/overview

3 Likes
3

Thanks for the reply :slight_smile:

It would have been great if we could use Custom GPTs via API, as it would allow more flexibility in integrating them into applications. But looks like I have to use Assistant AI for it now.

1 Like