We’re trying to develop an external communication to a MyGPT we created, however, we’re stuck on trying to locate the Endpoint as per OpenAI documentation.
As per documentation , it is stated :
Locate the Endpoint Information: Within the details of your MyGPT instance, there should be information regarding the API endpoint. This typically includes the base URL and may also specify the particular path or parameters needed for requests.
Anyone can point to the right direction?
Welcome to the forum!
Where did you find that information? google links back to your post 
If you looked at API docs, those pertain to the assistants API, not to the custom gpts.
Are you trying to create an action? I don’t think it’s possible to post information directly to a custom gpt, it has to be fetched.
XD I guess im the first to ask.
Where did I get that information? - I asked GPT4 " How do I find my endpoint for my MyGPTs? "
And it answered these :
To find the endpoint for your MyGPT instance in OpenAI's ChatGPT, you typically follow a series of steps within the OpenAI API dashboard or a similar interface provided by OpenAI. Here's a general guide on how you can locate your MyGPT endpoint:
1. **Log into OpenAI Dashboard**: Start by logging into your OpenAI account. You will need to access the API dashboard where your API keys and application settings are managed.
2. **Navigate to the MyGPTs Section**: Look for a section or tab related to MyGPTs. This is where you can manage your custom instances.
3. **Select Your MyGPT Instance**: If you have created multiple MyGPT instances, you'll need to select the specific one you want to use. Each instance may have its own unique endpoint.
4. **Locate the Endpoint Information**: Within the details of your MyGPT instance, there should be information regarding the API endpoint. This typically includes the base URL and may also specify the particular path or parameters needed for requests.
5. **API Documentation**: Refer to the provided API documentation for any additional details on how to structure your requests. The documentation will often include examples of how to use the endpoints with different programming languages and tools, such as `curl`.
I understand the part on how to send or make a request to ChatGPT based on the documentation : https://platform.openai.com/docs/api-reference/making-requests
However, instead of sending to
https://api.openai.com/v1/chat/completions
, I would like to send it to the apps I created , but I could not find the address or endpoint I should be sending my requests…
When you say
you’re talking about your gpts on chat.openai.com?
or are you talking about assistants?
or are you talking about your own webserver?
Also, don’t trust gpt-4, it hallucinates if you’re not careful.
My GPTs in chat openai com , yes.
I have created an Application there which referenced back to my PDF.
I would like to send a query or request to that Custom GPT instead of openai directly.
Oh, that’s not going to work, but you can use assistants for that if you’re in a pinch.
https://platform.openai.com/playground?mode=assistant
They’re very similar, but they can be accessed via the API.
nope, not currently.
you’ll have to use assistants or something. but they’re very similar
I can’t believe I’m recommending assistants
And its still in Beta Too!
Its okay then. Thank you very much for the help.
I’ll try to setup Assistants and see if it works with me. 
good luck!
Assistants aren’t an optimal solution, but they’re probably the easiest to just get something going with.
