When I generate an image using the OpenAI node in N8N, my original prompt gets automatically revised into a “better” or more detailed version.
The issue with this is that I want consistent visuals based on the prompt I provide. Each time I run the workflow, the prompt is revised differently, leading to inconsistent visuals.
Does anyone know how to configure the OpenAI node so that it uses my exact prompt, rather than revising it?
This is what the revised prompt looks like for more context:
It looks like your topic is missing some important information. Could you provide the following if applicable.
Hi @NickM ,
Welcome to the community! 
What version of n8n/OpenAI node are you on? I’ve recreated by don’t get the same output
Thank you for your response! I’m on the version 1.53.2 of n8n. I think you’re getting the download link for the image. You will get the same output as me, when you select the option ‘respond with image urls’.
I wasn’t getting roughly the same outputs on my prompts, so I started looking around a bit to see where I could find my given prompt.
When I selected, respond with image URL’s, the ‘revised_prompt’ showed up. I’ve searched the options, but I couldn’t get the Node to just use my original prompt.
