DALL-E 3 already available in OpenAI node

Describe the problem/error/question

Is the DALL-E 3 already available in OpenAI node? Because when I create an image the results are very different then the ones I create in GPT website.

What is the error message (if any)?

Please share your workflow

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Information on your n8n setup

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:

Hi @pehaem, looks like the node simply calls the /v1/images/generations endpoint provided by OpenAI without specifying a model:

From looking at the OpenAI API reference it appears they don’t allow specifying a DALL-E 3 as a model yet I am afraid: OpenAI Platform


I think this is a change that is super recent. But there is documentation on how to do DALLE-3 with curl, Python, or Node.js at the moment. I’d opt to use the curl if it’s supported by the javascript in n8n.


1 Like

Ah sweet! In this case you can simply send the request using the HTTP Request node like so:

You’ll get a result URL:

So depending on what you plan to do with the image you might need to add another HTTP Request node afterwards actually downloading the file generated by OpenAI.

When testing this just now I got a file like this fwiw:

Hope this helps :smiley_cat:


WOW! So cool!

So the image could be saved to a hard drive, or even posted to a Discord channel.

I’ve been cleaning up my Gmail accounts by posting the emails as archive to Discord and then deleting them from my email. And it started me thinking why not start my own little private Discord channel for DALLE3 images?

Thank You for putting that sample code together so quickly.

1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.