> ## Documentation Index
> Fetch the complete documentation index at: https://docs.jiekou.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Gemini 3 Pro Image Editing

Gemini 3 Pro Image (Nano Banana Pro) is designed to solve the most challenging image generation tasks by integrating state-of-the-art reasoning capabilities. It is the best model for handling complex and multi-turn image generation and editing, with higher accuracy and better image quality. It supports images encoded as URLs and Base64 as input.

## Request Headers

<ParamField header="Content-Type" type="string" required={true}>
  Enum value: `application/json`
</ParamField>

<ParamField header="Authorization" type="string" required={true}>
  Bearer authentication format: Bearer \{\{API Key}}.
</ParamField>

## Request Body

<ParamField body="size" type="string" default="1K">
  Pixel size of the output image. 1K (1024px), 2K (2048px), 4K (4096px). Defaults to 1K

  Optional values: `1K`, `2K`, `4K`
</ParamField>

<ParamField body="google" type="object">
  Google Search options

  <Expandable title="properties" defaultOpen={true}>
    <ParamField body="web_search" type="boolean" default={false}>
      Enable Google web search to generate more accurate images based on real-world information.
    </ParamField>
  </Expandable>
</ParamField>

<ParamField body="prompt" type="string" required={true}>
  Text prompt describing the desired image editing effect
</ParamField>

<ParamField body="image_urls" type="string[]">
  List of input image URLs for editing
</ParamField>

<ParamField body="aspect_ratio" type="string">
  Aspect ratio of the output image

  Optional values: `1:1`, `3:2`, `2:3`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`
</ParamField>

<ParamField body="image_base64s" type="string[]">
  List of Base64-encoded images for editing
</ParamField>

## Response Information

<ResponseField name="image_urls" type="string[]" required={true}>
  List of edited image URLs
</ResponseField>

<ResponseField name="grounding_metadata" type="object" required={false}>
  Model grounding metadata. Returned when Google server-side tools (such as Search) are called. Note: tool invocation depends on Google. Enabling the parameter does not guarantee that a tool call will be triggered 100% of the time.
</ResponseField>