CADABRA AI

AI images just changed: 5 upgrades, the models and every prompt.

1. The words spell right

GPT Image 2.6 plans before it draws, Ideogram 4.5 spells exactly what you put in quotes, and MiniMax H3 Image follows a long prompt with text inside the picture. Put every word in quotes and say where it goes.

2. One face, every picture

Nano Banana 2.1 takes up to 14 reference images, FLUX 3 up to 10 and MiniMax H3 Image up to 16. One clean character sheet, sent every time, keeps the same character in the rain, in a cinema, inside a wave and at the fridge.

3. Change one thing

Edit instead of regenerating: name the one thing to change and add keep everything else exactly the same.

4. Pictures come apart

Split into layers with Seedream 5.5: one picture becomes up to 20 transparent layers, with the hidden parts filled in, ready for Photoshop, Figma or After Effects.

5. Pictures move

An animated GIF from one sentence or from a picture in GPT Image 2.6, and a grey Blender blockout turned into a real house with FLUX 3.


Home · Pricing · What's new · All engines · All effects · Help center · Lessons · Meital Sela · AI video · AI images · AI voice · AI music · Image to video · What it costs · Credits, not a subscription · Blog · Connect to Claude and ChatGPT · Presets · Privacy · Terms

Cadabra AI · Meital Sela · Rehovot, Israel