I’m trying to map multiple images to ChatGPT Vision.
The number of images varies from run to run, that’s why I need it to be mapped to an array.
I’m pretty sure my structure is correct but yet it doesn’t work. Any ideas?
Hi @Krzysztof_L,
the structure looks good, except the comma and open curly bracket at the end. You should remove those.
The other possible issue is that the Google Drive links might not work as expected. I’ll check if that works real quick. If it doesn’t, I have a different solution.
Have you tried this with just one image to see if that would’ve worked?
Best,
Julius
I actually just worked on solving this for 3 hours straight. I was able to solve it. Works well now
Made a quick video for you:
Also, here is the blueprint to the working scenario:
blueprint (1).json (71.0 KB)
2 Likes
You are correct here, I was deleting part of the json and missed this one. However even when a proper structure was used it didn’t work.
Thank you so much for looking into this. I am recording a quick video reply too for you 
Did you watch the video? There were several problems with it that I all talked about in the video.
Try importing the blueprint and see if the scenario on the very bottom works (I added all approaches to it so you can easily see what I tried and why it didn’t work).
1 Like
Yes, I watched it throughout!
Thank you so much for recording this for me! I really appreciate it.
I didn’t see the notification about your reply, and I ended up finding a solution about one hour after posting here. It’s different from yours, and I thought I would record a video for the other way, too, so that maybe in the future, it will be helpful to someone.
2 Likes
Oh wow, this is great!
Didn’t know that was possible!
That’s a great solution. Well done! 
1 Like