ChatGPT interface timing out?

I am trying to use ChatGPT to interrogate a folder of pdf documents but every time I make a request it seems to time out (see screenshot). I have input my API code in settings and can perform the same query in ChatGPT itself (and other queries with no delay). What might be going on here?

Thanks.

Did you check your billing details and either add credits on your own or allow OpenAI to automatically handle this?

Im on a free plan but this seems to indicate that I have credits which I am not exceeding? If I add a query to the ChatGPT web interface then this is carried out normally?

The “free” plan is not unlimited.

Using the chat on the web is not the same as using the API. Did you read the Getting Started > AI Explained in the built-in Help and manual?

The free plan doesn’t include any API credits AFAIR.

I am new to AI but use the free version of ChatGPT on Safari web browser. So far my use has been limited, but I can see big potential for using Chat within DEVONthink to create annotations and summaries.
Does your comment “The free plan doesn’t include any API credits AFAIR” mean that I need to upgrade to a paid version of ChatGPT to use it within DEVONthink?
I will do that if necessary, but only if that is the reason ChatGPT is not working inside DEVONthink at the moment.

API credits do not require a paid subscription, this is independent. For more details see your OpenAI account and the DEVONthink 4 documentation.

My settings in DEVONthink seem to match guidance in the DT4 manual. ChatGPT settings show that the API key is being used, which suggests a valid connection between DT4 and ChatGPT. However, chats or summarise commands initiated in DT4 result in an “You exceeded your current quota, please check your plan and billing details” message. If I try the same chat with ChatGPT in the web browser it works. I am stuck.

And did you check the billing details of your account and add some credit for the API usage?

As an update, I have switched to using a Gemini API and this is working fine.

ChatGPT in the browser is not the same as accessing the LLM via the API. Did you read the Getting Started > AI Explained. section in the built-in Help and manual ?

I have read the help and the manual, and ChatGPT settings. My DEVONthink settings for AI seem to be correct.
ChatGPT shows that the API for DEVONthink is being used which infers there is a connection. However, I get an error message that I have exceeded quota, even though I have now upgraded to ChatGPT Plus.
If I use ChatGPT via web browser it works perfectly.
I do not know what to try next.

If I use ChatGPT via web browser it works perfectly.

Again, ChatGPT in the browser has nothing to do with what’s going on in DEVONthink.

I do not know what to try next.

Did you give OpenAI money to use their API? You need to go into the API account and make sure there is money to spend against, e.g., putting in $20.

Thank you. Problem solved. AI Now working.
I am new to AI. Despite reading the help and manual I had not fully understood the distinction between payment to Open AI and ChatGPT Plus.
I appreciate your patience and steering me in the right direction to resolve the issue.

1 Like

You’re very welcome.
I will look at what I wrote and see if I can make the distinction more clear.

Cheers!