AI Image Generator. Your idea, in the frame.

Describe a product shot, sketch a new world, or rework a photo. Create and edit with GPT Image and Nano Banana models, all in one workspace.

10

Create · reference · refine

One workspace for the first idea and the next revision

Start with words when you need a new direction. Bring in a reference when you already know the subject, look, or layout you want to explore.

01

Turn a brief into an image

Describe the subject, setting, style, and light. Build product visuals, illustrations, or concept art from a text prompt.

02

Put your references to work

Upload a photo or sketch in image-to-image mode. Tell the model which details to keep and which parts to change.

03

Choose the model for the job

Start with GPT Image 2.5 or explore GPT Image 2 and the Nano Banana family. Compare different approaches to the same brief.

04

Fit the place it will live

Set the aspect ratio and an available resolution before generating. Make a square post, a portrait campaign, or a wide storyboard frame.

What you can make

From product launch to story pitch

Explore three creative directions: a campaign built around one product, a sequence of story frames, and a new take on an existing design.

Amber serum bottle in coordinated blue campaign layouts
Product campaigns

One product. More ways to show it.

Use a product reference to explore lighting, backgrounds, and compositions for your next campaign.

Four rainy-city storyboard panels with a recurring character
Storyboards & concepts

Give the story its first frames

Describe the character, location, and mood, then develop a visual direction for each scene.

Sneaker concept before and after a blue material edit
Image editing

Find the next version of your design

Try a different color, material, or setting. Specify the details you want to carry into the new image.

How it works

Your next image starts with a clear brief

Write the idea, choose your settings, and generate. Each result gives you something concrete to refine.

  1. 01

    Describe the image

    Name the subject and the intended use. Add the composition, style, lighting, and any text you want in the scene.

  2. 02

    Add a reference

    For image-to-image creation, upload a photo or sketch and explain what it should guide.

  3. 03

    Set model and format

    Choose an image model, aspect ratio, and available resolution. Check the credit cost shown before generating.

  4. 04

    Generate and refine

    Sign in to generate with your credits. Review the result and use it as a reference for a more focused revision.

A better starting prompt

Describe the picture, then the details

A useful prompt tells the model what to show and how to frame it. If you upload a reference, give it a clear role.

Product photo prompt

Create a studio product photo of the amber serum bottle in the reference. Place it on a cobalt-blue plinth against a clean blue background. Use soft side lighting and realistic reflections on the glass. Keep the bottle silhouette and label placement; leave space on the left for a headline. Vertical 4:5 composition.

  • Subject

    Name the main object and the details that define it.

  • Composition

    Explain the camera angle, framing, and space for copy.

  • Look & light

    Choose a palette, style, materials, and light direction.

  • Reference role

    Say what to preserve and what you want to change.

AI image generator FAQ

Before you make your first image

Answers about prompts, references, formats, and credits.

How does the AI image generator work?

Choose an image model, write a prompt, and set the format. The model generates a new image from your description and any references you add.

Can I create an image using only text?

Yes. Choose text-to-image mode and describe the subject, scene, style, and lighting. You do not need a reference image.

Can I edit or restyle my own photo?

Yes. Use image-to-image mode with a supported model, upload your photo, and describe the changes. Available file types and reference limits are shown in the workspace.

Which image models are available?

This page starts with GPT Image 2.5. The model menu also includes GPT Image 2, Nano Banana 2, Nano Banana 2 Lite, Nano Banana Pro, and other available image options.

What image sizes can I choose?

Aspect ratios and resolutions depend on the selected model. The workspace shows the options it supports, so you can choose a format for social posts, product pages, or wider compositions.

Do I need an account or credits?

You need to sign in and have enough credits to generate. The workspace displays the cost for your settings. Daily check-in rewards can add credits; downloading generated images requires a paid plan.

Make the first version

The next image starts with your idea

Write a brief or bring a reference. Choose your image model and start creating right here.