Hi everyone,
I’ve been experimenting with automating some AI-based image tasks inside n8n and I’m looking for advice from others who may have tried something similar.
My goal is to build a workflow that can:
-
send an image to an external AI editing service
-
generate or modify an image based on prompts
-
return the edited output back into the workflow
-
store or forward the result automatically
Recently I’ve been testing Nano Banana (https://nano-banana.io/), which provides prompt-driven image editing and generation. The results are pretty good, but I’m still figuring out the best way to integrate it smoothly into an n8n pipeline—especially around handling API responses and file nodes.
Has anyone here set up image-editing or image-generation pipelines in n8n before?
I’d love to hear how you structured your workflow, which nodes you used, or any performance tips you might recommend.
Thanks in advance!