Why Am I Unable to Read an Image Using Google Gemini Despite Successful Trial with GPT4?

Bugs
1

This topic was automatically generated from Slack. You can find the original thread here.

Hi sorry in advance if here is not the place for post my issue:

I am trying to read an image using google gemini. But I am getting:

ENOENT: no such file or directory, open

I tried the:

/tmp/{{mysteps}}
tmp/{{mysteps}}
{{mysteps}}

So, do not know really how to get the image.

Thanks for your help.

Edit: I used GPT4 for the same purpose with no issues.

2

can you help?

3

It seems like your question is incomplete. Could you please provide more details or clarify what specific help you need with Pipedream?

4

I am sending a picture using webhooks. I can see the URL of the picture in the path. But when I try to generate text from the image using google gemini. I am getting an error of the path location. This is a screenshot of the error:

Screenshot 2024-02-06 205008
Screenshot 2024-02-06 2050081120×721 36.7 KB

As I mentioned above, I tried different paths to get the tmp folder, but really don’t know what I am doing wrong.

/tmp/{{steps.trigger.event.body["0[0][guid]"]}}
tmp/{{steps.trigger.event.body["0[0][guid]"]}}

In the message error I got the URL of the image inside [] brackets. Don’t know if this helps.

Google Gemini Account
Google Gemini
Prompt Text
What is this picture about?
MIME Type
image/png
Image File Paths
[ "tmp/https://......png" ]