AI models · OpenAI
GPT Image 2 online
Use GPT Image 2 in your browser with Omni Banana to create images from a description or edit your photos. GPT Image is OpenAI’s family of image models, the ones behind image creation in ChatGPT. They follow long, detailed prompts closely and handle text inside images well.
What GPT Image 2 can do
| Made by | OpenAI |
|---|---|
| Create from a description | Yes |
| Edit your photos | Yes, up to 4 photos |
| Aspect ratios | 1:1, 3:2, 2:3 |
| Resolution | Standard |
| Quality levels | low, medium, high |
| Longest prompt | 5,000 characters |
How to use GPT Image 2 in Omni Banana
- Open the studio and sign in with Apple or Google.
- GPT Image 2 is already chosen when you use the button on this page.
- Describe the image you want, or add up to 4 photos and say what to change.
- Pick the aspect ratio and the number of images, then press Generate.
Or start from one of the 140+ templates, such as a business portrait or a photo restoration, and choose GPT Image 2 as the model.
Other OpenAI models
Compare with other models
- Nano Banana 2 by Google
- Nano Banana Pro by Google
- Seedream V4.5 by ByteDance
- Flux Kontext Pro by Black Forest Labs
- All 31 models side by side
Questions
Is this the official GPT Image 2?
GPT Image 2 is made by OpenAI. Omni Banana gives you access to it through hosted AI providers, next to 30+ other models. Omni Banana is not affiliated with OpenAI.
Can GPT Image 2 edit my own photos?
Yes. Add up to 4 photos in the studio and say what to change. You can also create an image from a description alone.
How much does GPT Image 2 cost?
Each image uses credits, and the studio shows how many before you generate. Credits come from a weekly or monthly plan or a one-time pack, and never expire.