Create images from a prompt or rework your own pictures with Nano Banana 2.1, Nano Banana 2 Lite and GPT Image 2.5, all in one place.
Four image models share one form, so you can switch between them without retyping your prompt.
Describe the picture you have in mind and get it back as an image. Say what the subject is, where it is, and what the light and style should look like.
Upload your own pictures and tell the model what to change: swap the background, restyle a product shot, or combine several references into one scene.
Google's Nano Banana 2 Lite is the quick 1K option and Nano Banana 2.1 the full model up to 4K. From OpenAI, GPT Image 2.5 Flare is the faster model and Sunburst the higher-quality one.
Nano Banana 2.1 and both GPT Image 2.5 models offer 1K, 2K and 4K output. Start at 1K while you work out the prompt, then raise the resolution for the final image.
Every model covers square, landscape and vertical formats up to 21:9. The Nano Banana models add panoramic 4:1, 1:4, 8:1 and 1:8, and Auto lets the model choose.
Ask for one to four images from a single prompt and compare the variations side by side.
It runs in your browser. There is nothing to install.
Choose Text to Image to start from a prompt, or Image to Image to work from your own pictures. Then select Nano Banana 2 Lite, Nano Banana 2.1, GPT Image 2.5 Flare or GPT Image 2.5 Sunburst.
Describe the subject, the setting and the style. In Image to Image, upload your reference pictures and say what should change.
Choose the aspect ratio, the resolution where the model offers one, and how many images you want. The credit cost updates under the button before you generate.
Results appear in the panel on the right. Click an image to view it full size, download it, or find it again later under My Assets.
Something we missed? Write to support@seedance.kr.
Nano Banana 2 Lite and Nano Banana 2.1 from Google, and GPT Image 2.5 Flare and GPT Image 2.5 Sunburst from OpenAI. Lite is built for fast 1K drafts and 2.1 is Google's full model. Flare is the faster GPT Image 2.5 model and Sunburst the higher-quality one, at the same credit cost. All four support both text-to-image and image-to-image.
New accounts receive free credits on sign-up, and you can spend them on images. Each image costs credits depending on the model and resolution, and the exact cost is shown under the Generate button before you start.
In Image to Image you can upload up to 10 reference images with the Nano Banana models and up to 16 with the GPT Image 2.5 models. Each file can be up to 10MB.
All models offer 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9 and Auto. The Nano Banana models add 5:4, 4:5, 4:1, 1:4, 8:1 and 1:8, and the GPT Image 2.5 models add 27:16, 16:27, 9:8 and 8:9 at 1K only. Nano Banana 2.1 and GPT Image 2.5 output 1K, 2K or 4K; Nano Banana 2 Lite always outputs about 1K.
Every image is saved to your account. Open Assets from the user menu to view, download or delete them.
Yes. Download the image, then upload it under Image to Video on the home page and animate it with Seedance, Veo or Grok Imagine.
No. Seedance.kr is an independent service. Images are generated with Google's Nano Banana models and OpenAI's GPT Image 2.5 models.
Free credits land in your account as soon as you sign up. No card, nothing to install.