Lummi

Lummi works differently from a prompt-box image generator. Instead of asking you to type a description and wait for a render, it gives you a browsable library of AI-made visuals split into three tabs: Photos, Illustrations and 3D. The homepage sums it up as "powered by humans and AI", and that's the mechanic. Images are published under named creators, with credits like Pablo Stanley, Clemara Studio, Semihan Dosunmu and Sanna Granqvist showing up in the Daily picks feed. Some of that work carries a Pro badge, so the free library and the paid one aren't identical.

On top of the library sits a small set of editing tools. The site lists three: Reframe, Restyle and Background removal. In practice that means you find something close to what you need, then adjust it instead of starting from zero. There's also a ⌘K search shortcut, which tells you the team expects you to search fast and often, the way you'd use Unsplash rather than the way you'd use Midjourney.

The pitch is "no more boring design": stock visuals that don't look like the same five handshake photos everyone else uses. It's freemium, with a free library to browse and paid plans listed from $8/mo, plus a "Try Pro for free" trial on the homepage. The honest caveat is that the public page is light on detail. It names the tools but doesn't explain how far Reframe or Restyle go, and licensing terms aren't spelled out where we looked. If you need quick, slightly unusual hero images, illustrations or 3D renders for a landing page or deck, the model makes sense. If you need precise control over a specific scene, you'll want a true generator alongside it.