I can upload images to the GPT, but I can’t get it to display those images to the user. Am I doing it incorrectly?
For context, I’m trying to make a little text-based roleplaying game. I want to be able to show the user a map of the dungeon. I can upload the map to the GPT, but when I tell the GPT to display the image, it won’t.
I’m having the same issue, I even tried to use some sort of markdown instruction, but it doesn’t seem to work, any ideas?
I have a feeling the images aren’t supposed to be displayed.
I’ll let you know if I figure something out.
My thought too. Will likewise let you know if I stumble upon anything.
Okay, so, it can work, kind of.
I’ve got it displaying images. Sometimes it displays the image as intended. Sometimes it gives a download link. Sometimes it glitches.
In my context, where I’m trying to be obtuse about it, creating a sort of illustrated dungeon crawl, it doesn’t work very well. The GPT prefers more direct instructions, I think. If the user says: “show me image.jpg, it will often try.”
If I give it behind-the-scenes instructions to show the image at a certain point in the story, and the user triggers it without asking for it directly, it doesn’t work as well. Maybe 25% of the time.
Now it’s working consistently. I’m not sure what changed. I can even put it in the custom instructions part now. I say: show the user image.jpg, and it does.
It uses Code Interpreter to do it. I’m not sure if that has something to do with it.
Yeah, seems to work for me now too… I turned on code interpreter and that did help. I’m hoping it consistent, but thanks for your help.
Correct, you do need code interpreter enabled to retrieve uploaded files