I built Blooming, a visual AI workspace where you drag-and-drop “nodes” (text, image, video) onto a whiteboard and chain AI text, image and video models together.

It's kinda like n8n but for AI art:

• Node-based canvas

• Multi-model switching (test different models side by side and iterate multiple versions)

• Pipe text or image output to other nodes to refine prompts, turn them into videos or explain images

Demo link: https://youtu.be/TdFzhxeRFNg

I’d love your feedback on UI clarity, performance, and which models to integrate next.

What’s confusing or missing for AI image and video power users?

Everything here is live in prod (you can sign in to use it). Trying to see if this is useful to people.

For those interested in the tech stack I used see comment below

Thanks for checking it out – Edrick