AI Image and image models
AI Image creates new images from text and edits or combines images you upload.
Text to Image and Image to Image
- Text to Image — describe the picture you want.
- Image to Image — upload one or more images and describe what to do with them: change the style or background, put a product in a new scene, combine people or objects from several pictures, and more.
You can create up to 4 images per request. The price is per image, so 4 images cost four times as much as one.
Choosing a model
| Model | Good to know |
|---|---|
| Nano Banana | A good all-round choice for creating and editing. |
| Nano Banana 2 / Nano Banana Pro | Higher resolutions up to 4K. Up to 14 reference images. |
| GPT Image 2 / GPT Image 2.5 | Choose resolution (1K–4K) and quality level; higher quality costs more. |
| Seedream 5.0 Lite / Pro | Accepts up to 10 reference images for editing and combining. |
| FLUX.2 Pro, Grok Imagine Image | Alternatives with their own looks. |
| FLUX.1 Schnell | Text to Image only. The lowest-cost option for quick drafts. |
The app always shows the current models, options and prices.
Using Characters
Characters is a library of ready-made people you can use again and again, so the same person appears across your images and videos. Open a character and choose Create a scene (image) or Create a video. There is no extra charge for using a character; you pay the normal price of the model.
Prompt tips
- Describe the subject first, then the setting, style, lighting and camera angle.
- For edits, say exactly what should change and what should stay the same.
- Use the Optimize button to have AI improve your prompt.