GET

/v1/schemas/fal/image/fal-ai/nano-banana-2/edit

Accept:

kind inputactivity imagehash 0c071689…extracted Nov 10, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `be07769b1fb1174422689255c4ba7ed491a0a1991f6b3de4e39e59b09e1272a4` { "fetchedAt": 1794286801, "extractorVersion": "2", "verifiedAt": null }

Request body

24 properties · 1 required · 7 $defs

property	type	description	constraints
prompt*	string	The prompt for image editing.	length ≥ 3 · length ≤ 50000
safety_tolerance	enum	The safety tolerance level for content moderation. 1 is the most strict (blocks most content), 6 is the least strict.	1 \| 2 \| 3 \| 4 \| 5 \| 6 · default: "4"
enable_image_search	boolean	Ground image generation with Google Image Search. Can be combined with enable_web_search; source attribution is returned in grounding_metadata. Uses direct Google or Vertex APIs.	default: false
num_images	integer	The number of images to generate.	1 ≤ n ≤ 4 · default: 1
system_prompt	string	Optional system instruction that steers the model's persona and output style across the request. Leave blank to omit; when provided, it is sent as the system instruction to Gemini (or as a system message on OpenAI-compatible providers).	length ≤ 50000 · default: ""
audio_url	string	Optional audio file to use as input context. Accepts an http(s)/data URL (downloaded and sent inline, max 15MB).	nullable
thinking_level	enum	Controls thinking effort ('minimal' or 'high') and requests thought summaries. Omit to use the provider default ('minimal'); thinking cannot be disabled.	minimal \| high · nullable
top_p	number	Nucleus sampling probability. Omit to use the provider default of 0.95. Explicit values use a direct Google or Vertex API.	0 ≤ n ≤ 1 · nullable
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
seed	integer	The seed for the random number generator.	nullable
include_thoughts	boolean	Whether to include thought summaries. Omit to preserve the existing behavior: an explicit thinking_level requests summaries. Set limit_generations=false to return summary text in description. This does not disable thinking or return intermediate thought images.	nullable
video_url	string	Optional video to use as input context. Accepts an http(s)/data URL to a video file (downloaded and sent inline, max 15MB) or a YouTube URL (passed through to the model without downloading).	nullable
context	GeminiTextPart \| GeminiImagePart \| GeminiDocumentPart[]	Ordered text, image, PDF, UTF-8 text-file or video parts appended to the current prompt. Original media bytes are preserved; images count toward the total 14-reference limit.
max_output_tokens	integer	Maximum output tokens for the generation. Omit to retain the provider default. Explicit values use a direct Google or Vertex API.	1 ≤ n ≤ 32768 · nullable
history	GeminiVideoMessage[]	Previous native Gemini turns, including original parts and thought signatures from conversations. The current prompt starts a new user turn.
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 and ignore any intermediate images generated by the model. This…	default: true
output_format	enum	The format of the generated image.	jpeg \| png \| webp · default: "png"
enable_web_search	boolean	Enable web search for the image generation task. This will allow the model to use the latest information from the web to generate the image.	default: false
temperature	number	Sampling temperature. Omit to use the provider default of 1. Explicit values use a direct Google or Vertex API.	0 ≤ n ≤ 2 · nullable
aspect_ratio	enum	The aspect ratio of the generated image. Supports extreme ratios: 4:1, 1:4, 8:1, 1:8.	auto \| 21:9 \| 16:9 \| 3:2 \| 4:3 \| 5:4 · +9 more · nullable
pdf_url	string	Optional PDF document to use as input context. Accepts an http(s)/data URL (downloaded and sent inline, max 15MB).	nullable
return_history	boolean	Return reusable native conversations with original image parts and provider thought signatures. Uses a native provider. Each generated conversation is independent; send its messages as history for the next turn.	default: false
resolution	enum	The resolution of the image to generate.	0.5K \| 1K \| 2K \| 4K · default: "1K"
image_urls	string[]	The URLs of the images to use for image-to-image generation or image editing. Optional when at least one of `video_url`, `audio_url`, or `pdf_url` is provided.

Types

7 $defs

GeminiAudioPartobject

thought	boolean	Provider thought marker, preserved when replaying a returned conversation.	nullable
type*	const	The audio part discriminator.	"audio"
audio_url*	string	Original audio URL or data URI from a previous Flash Image request (up to 15 MB).
thought_signature	string	Opaque provider-issued thought signature. Replay it unchanged on its original part.	length ≥ 1 · nullable

GeminiDocumentPartobject

thought_signature	string	Opaque provider-issued thought signature. Replay it unchanged on its original part.	length ≥ 1 · nullable
thought	boolean	Provider thought marker, preserved when replaying a returned conversation.	nullable
file_url*	string	PDF or UTF-8 plain-text document URL or data URI (up to 15 MB).
type*	const	The document part discriminator.	"document"
mime_type*	enum	The document's media type.	application/pdf \| text/plain

GeminiImagePartobject

image_url*	string	Original image URL or data URI. Replayed without resizing or re-encoding.
thought	boolean	Provider thought marker, preserved when replaying a returned conversation.	nullable
type*	const	The image part discriminator.	"image"
thought_signature	string	Opaque provider-issued thought signature. Replay it unchanged on its original part.	length ≥ 1 · nullable

GeminiSignaturePartobject

thought	boolean	Provider thought marker, preserved when replaying a returned conversation.	nullable
type*	const	The signature part discriminator.	"signature"
thought_signature*	string	Provider-issued signature-only part, replayed unchanged.	length ≥ 1

GeminiTextPartobject

thought	boolean	Provider thought marker, preserved when replaying a returned conversation.	nullable
type*	const	The text part discriminator.	"text"
text*	string	Text at this position in the conversation.
thought_signature	string	Opaque provider-issued thought signature. Replay it unchanged on its original part.	length ≥ 1 · nullable

GeminiVideoMessageobject

role*	enum	Native Gemini conversation role.	user \| model
parts*	GeminiTextPart \| GeminiImagePart \| GeminiDocumentPart[]	Ordered text, image, document, video, audio, and signature parts.	items ≥ 1

GeminiVideoPartobject

thought	boolean	Provider thought marker, preserved when replaying a returned conversation.	nullable
type*	const	The video part discriminator.	"video"
video_url*	string	Video URL or data URI (up to 15 MB and 120 seconds), or a YouTube URL. Supported by Flash Image only.
thought_signature	string	Opaque provider-issued thought signature. Replay it unchanged on its original part.	length ≥ 1 · nullable

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fal","endpointId":"fal-ai/nano-banana-2/edit","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}