fal.ai·Image generation

Nano Banana

/fal-ai/nano-bananaasync$0.0796 / image

Google's famous original image generation and editing model Returns hosted file URLs (download promptly; fal keeps them for a limited time). Billed per image.

Charged per image produced; a call that produces nothing is charged one unit.

Input

aspect_ratiostring
The aspect ratio of the generated image. (default "1:1")
21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, …
num_imagesinteger
The number of images to generate. (default 1)
sync_modeboolean
If `True`, the media will be returned as a data URI and the output data won't be available in the request history. (default false)
safety_tolerancestring
The safety tolerance level for content moderation. 1 is the most strict (blocks most content), 6 is the least strict. (default "4")
1, 2, 3, 4, 5, 6
output_formatstring
The format of the generated image. (default "png")
jpeg, png, webp
promptstringrequired
The text prompt to generate an image from.
seedany
The seed for the random number generator.
limit_generationsboolean
Experimental parameter to limit the number of generations from each round of prompting to 1. Set to `True` to to disregard any instructions in the prompt regarding the number of images to generate. (default true)

Field	Type	Required	Description
aspect_ratio	string (21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, …)	—	The aspect ratio of the generated image. (default "1:1")
num_images	integer	—	The number of images to generate. (default 1)
sync_mode	boolean	—	If `True`, the media will be returned as a data URI and the output data won't be available in the request history. (default false)
safety_tolerance	string (1, 2, 3, 4, 5, 6)	—	The safety tolerance level for content moderation. 1 is the most strict (blocks most content), 6 is the least strict. (default "4")
output_format	string (jpeg, png, webp)	—	The format of the generated image. (default "png")
prompt	string	yes	The text prompt to generate an image from.
seed	any	—	The seed for the random number generator.
limit_generations	boolean	—	Experimental parameter to limit the number of generations from each round of prompting to 1. Set to `True` to to disregard any instructions in the prompt regarding the number of images to generate. (default true)

Run it

curl -s https://sooma.dev/v1/run \ -H "Authorization: Bearer $SOOMA_API_KEY" \ -H "Content-Type: application/json" \ -d '{"provider":"fal","endpoint":"/fal-ai/nano-banana","input":{"prompt":"A cinematic photo of a lighthouse at dawn, soft light"}}'

Or ask your agent: it will call inspect for this schema and run it over MCP.