You describe something in ordinary words. A few seconds later you have a picture of it. That is the whole idea, and everything else on this page is a variation on it.
You do not need to be able to draw, use design software, or know anything about how any of this works underneath.
Pictures from a description
Type what you want to see — a red fox asleep on a stack of library books — and you get an image of it. Not a photograph of a real fox; a new picture made to match your description.
You can ask for a style as easily as a subject: an oil painting, a pencil sketch, a product photo on a white background, a 1970s film still. See writing a prompt.
Pictures from other pictures
You can start from an image instead of a blank page:
- Give it a photo and a description, and get a new version in a different style
- Give it a photo of a person or product and keep that subject consistent across many images
- Combine a reference picture with a written description of what to change
This is covered in starting from an image.
Video
Short video clips, from a description or from a still image you want to bring to life. Video takes longer and costs more than an image, and there are a few extra rules about it — see making videos.
The same character, every time
Describing a person in words gives you a different person in every picture. An avatar is a character the models already know, so you get the same one repeatedly — across a set, a story, a product range.
You can claim one, or train your own from photographs. It also unlocks Apex Image, the best image model here, which does not run without one. See avatars.
Fixing and improving what you have
Once you have an image, you can:
- Remove something from it — Magic Eraser: paint over an object and it goes away, with the background filled in behind it
- Change one feature — Apex Edit: hairstyle, outfit, pose, makeup, lighting and more, leaving the rest of the picture alone
- Move a feature between pictures — Apex Transfer: a tattoo or an outfit
- Make it bigger and sharper — Apex Upscale, useful before printing
- Take a still frame out of a video — Frame Extractor
Asking what is in a picture
Apex Analyze works the other way round: give it an image or a clip and it describes what it sees, in words.
Useful when you want to work out how to describe something you like, or need a caption. The result is text rather than a picture.
Doing it many times over
If you need thirty product shots rather than one, you do not have to sit and click thirty times. Bulk generation runs a batch and tells you when it is done. See generating in bulk.
What it costs
Generating spends credits. Your plan includes an amount of them; different kinds of work cost different amounts, with video costing more than images. Browsing, organising and downloading your own work are all free.
The whole system is explained in how credits work.
What this is not
Worth saying plainly, because it saves disappointment:
- It does not search the internet for existing pictures. Everything is made new.
- It is not a photo editor with layers and brushes. You describe changes rather than making them by hand.
- It cannot reliably produce exact text inside an image, precise counts of objects, or an exact likeness of a specific real person.
- The same description twice will not give you the identical picture, unless you deliberately reuse the seed — see settings explained.
Ready?
Create your account, then make your first image.