Sending images to Open AI GPT 4o

Questions
,
1

:bullseye: What is your goal?

My goal is to send multiple images to GPT-4o so that he can analyse them and get some more context for its global task of summing up messages.
Thanks in advance for taking the time to read and help me!

:thinking: What is the problem & what have you tried?

The issue is that when I send files, I get the following error “Run has been ended with an invalid status “failed”. Sorry, something went wrong.”
I use the “Message an assistant” module and map the files through a simple array containing all the image URLs ([url1, url2,…]) I got from Airtable.
I’ve also tried to hardcode the url but it did not work out.
In the screenshot section, you can find the inputs I use.

:camera_with_flash: Screenshots (scenario flow, module settings, errors)

inputs.png
inputs.png384Ă—291 6.75 KB

3

Hey there,

which exact model is your assistant using? You can check here if its vision compatible:

Also chatgpt-4o-latest is getting deprecated in 5 days.

4

Hi,

Thanks for the reply !

I currently use GPT 4o system. This model should support vision.

I tried yesterday and it was working and today it wouldn’t work.

5

Yeah I suggest you switch to something else entirely, most of 4o is getting deprecated over the next couple of weeks.

1 Like
6

Okay, didn’t know that ! Thanks for the advice, I’ll try to use a more recent model