AI image generator
Describe an image in plain words, choose a style, and generate it in seconds.
Your image will appear here
Describe what you want to see, choose a style, and press Generate.
- Name the subject and the setting
- Mention lighting, mood and colors
- Add a camera, lens or style detail
How to use the AI image generator
This text-to-image AI turns a written description into a brand-new picture. No design skills or software needed — just a clear prompt.
- 1
Describe your image
Type what you want to see in plain language. Mention the subject first, then where it is and what the light is like.
- 2
Pick a style and aspect ratio
Choose a look such as Photorealistic or Cinematic, and a shape that suits where the image will be used: square for posts, 9:16 for stories, 16:9 for banners.
- 3
Generate and download
Press Generate and your image usually appears in about 5 seconds. Download it, copy the prompt, or adjust a detail and try again.
Prompt tips for better AI photos
The more specific your description, the closer the result. A strong prompt usually covers these six things.
- Subject
- Be concrete: “a ceramic espresso cup”, not “a cup”.
- Setting
- Place it somewhere: “on a marble café counter in Lisbon”.
- Lighting
- Light sets the mood: “soft morning window light”, “neon at night”.
- Camera
- For photos, name a lens or angle: “85mm portrait”, “top-down flat lay”.
- Style
- Add a look: “editorial magazine photo” or “minimal studio product shot”.
- Details
- Colors, materials and textures make results feel real: “linen, oak, brushed brass”.
Need inspiration? Browse example prompts.
Questions about the AI image generator
Quick answers for when you're generating. Plans and accounts are covered on the pricing page.
What can I make with this AI image generator?
New images from a description: realistic photos of places, food, products and invented people, and illustration-style images too. Our guides show real results for social media, blog and stock photos and food. It doesn't edit photos you upload; text-to-image vs image-to-image explains the difference.
Which style and aspect ratio should I start with?
Start with Photorealistic, the default, and pick the shape for where the image will go: 1:1 for square posts, 4:5 for feed posts, 9:16 for Stories and Reels, 16:9 for headers and slides. AI photo styles compared and the aspect ratio guide go further.
Can I make more than one image from a prompt?
On the free plan each prompt makes one image. Paid plans can make up to 4 at once, and each one uses one of your images; see pricing.
Why was my prompt refused?
We automatically refuse some prompts that break our Acceptable Use Policy, and our image provider applies its own content policy. A refused prompt doesn't count against your images. If a harmless prompt is flagged, try describing it differently.
What if a generation fails or takes too long?
Press Try again: retrying the same request in this browser is free, and if paid images never arrive they're returned to your balance within a day; see the pricing FAQ.
How do I keep the images I make?
Download each one as a PNG. Your recent images and prompts are also saved in this browser, but image links can expire, so download anything you want to keep.
Why doesn't the image match my prompt exactly?
The model fills in everything the prompt leaves out, and every generation is new. Name the details you care about, leave words out of the picture and try a couple of versions; why the same prompt gives a different image and how to make AI photos look real show how.