Nano Banana Pro Edit

Nano Banana Pro Edit · Google — Image to image · Up to 10 source images · Up to 4K. Call it through one queue API and pay for the options you pick.

modellane/nano-banana-pro-edit · Image to image · Google

Image to image model: up to 10 source images, up to 4K, 10 aspect ratios.

Vendor
Google
Task
Image to image
Output
Image
Price
From $0.19 per run

Input parameters

  • Prompt prompt — required Describe what you want produced as concretely as you can.
  • Source images images — required, range: 1–10 Images used in the generation; their order can affect the result.
  • Aspect ratio aspect_ratio — optional, options: 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9 Sets whether the output is square, portrait, or landscape.
  • Resolution resolution — optional, options: 1K, 2K, 4K The output resolution.
  • Web search enable_web_search — optional, default: No When on, the model searches the web while answering the request.
  • Output format output_format — optional, options: Default, PNG, JPEG The file format of the output.
  • Input media resolution media_resolution — optional, options: Default, Low, Medium, High The resolution the input media is processed at; a higher value can take longer.
  • Top-p top_p — optional, advanced, default: 0.95, range: 0–1 A sampling cutoff that limits output variety; a lower value gives a more consistent result.
  • Temperature temperature — optional, advanced, default: 1, range: 0–1 A higher value gives a more creative result, a lower value a more consistent one.
  • Seed seed — optional, advanced, default: Random The same seed and the same settings produce a similar result; leave it empty and one is picked at random.

Output and retention

The generated file and your input are kept for seven days and then deleted.

Queue API

Submit to the queue: POST /v1/queue/modellane/nano-banana-pro-edit

Status, result, cancellation and error codes work the same for every model; see the API docs.

Playground

Everything on this page is public; you need an account to run the model.