Hi everyone,
I’m exploring ways to bring some AI-powered image editing features into my n8n workflows, and I’m hoping to get some suggestions from people who’ve tried something similar.
The idea is to create a workflow that can take an input image, send it to an external AI tool for modifications (like background edits, object removal, or prompt-based changes), and then return the final result back to n8n for further steps.
I’ve been testing a tool called Nano Banana (https://nano-banana.io/), which offers prompt-based image editing and generation. It works nicely on its own, but I’m still figuring out the smoothest way to wrap it into an automated pipeline—especially when dealing with image uploads and handling the returned files.
If anyone has experience integrating AI image services into n8n, I’d love to hear how you approached it—what nodes you used, how you handled the API calls, and any pitfalls to watch out for.
Appreciate any insights!
