Hey all, been stuck on this for a few hours so hoping for another set of eyes.
Goal: user uploads an image, I send that image to OpenAI Vision API, it analyzes it and sends back a response that I display to the user.
Where I’m at:
I successfully “initialized” the OpenAI API.
I can send a hard-coded image URL (i.e. I copy/paste the image URL and send it via the API) and it works fine. I get a response back from OpenAI and see it populate in the database.
Where something is wrong is when I pass a user uploaded image into the backend workflow connected to OpenAI API. The image is being added to my Bubble database, but I am not received anything back from OpenAI. (Attached screenshots of my workflows)
I tried a few combinations of adding “https://” and removing JSON safe with no luck. I’ll keep trying that though because I may have missed something.
Can you point me in the direction of where I can learn more about “authenticating” a file so it’s visible to OpenAI? I currently have the data type of Background Image set to public with no privacy rules applied.