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

# Gemini 3 Pro Image VIP Image Generation

> Gemini 3 Pro Image VIP supports text-to-image and image-to-image generation with up to 14 reference images.

<Note>
  **Note for users in mainland China:** Please use `https://toapis.cn` as the API endpoint (Base URL). Replace `https://toapis.com` with `https://toapis.cn` in the examples in this document.
</Note>

## Version Options

| Version                                               | Reference image limit | Use case                                                        |
| ----------------------------------------------------- | --------------------- | --------------------------------------------------------------- |
| [Standard](../gemini-3-pro-image/generation)          | 6                     | Text-to-image and edits with a small number of reference images |
| [VIP](../gemini-3-pro-image-vip/generation)           | 14                    | Edits and compositions that need more reference images          |
| [Official](../gemini-3-pro-image-official/generation) | 14                    | When you need native generation parameter control               |

The reference image count is the total number of input images; it does not represent the number of output images. The three versions use different model IDs, so follow the parameters and examples on the corresponding page.

## Current Version

Use `model: "gemini-3-pro-image-preview-vip"` to select VIP, which supports text-to-image and image-to-image or image editing with up to 14 reference images.
It suits edits or multi-image compositions that need 7 to 14 reference images. In multi-image requests, you can explain the purpose of each image in list order.
Requests run asynchronously; after a successful submission, query the result by task ID.

<Warning>
  `image_urls` only supports image URLs and does not accept base64 directly. First use the [Upload Image API](../../uploads/images) to get an accessible URL.
</Warning>

## Authorizations

<ParamField header="Authorization" type="string" required>
  Authenticate with `Bearer YOUR_API_KEY`. Create an API Key in the [console](https://toapis.com/dashboard/tokens).
</ParamField>

## Body

<ParamField body="model" type="string" default="gemini-3-pro-image-preview-vip" required>
  Fixed to `gemini-3-pro-image-preview-vip`. The standard model name and other aliases do not automatically switch to this VIP route.
</ParamField>

<ParamField body="prompt" type="string" required>
  Describe the image to generate or edit. When you pass multiple reference images, you can explain the purpose of each image in list order.
</ParamField>

<ParamField body="size" type="string">
  Image aspect ratio, for example `1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`.
</ParamField>

<ParamField body="n" type="integer" default={1}>
  Generates 1 image per request. Use the number `1`, not the string `"1"`.
</ParamField>

<ParamField body="image_urls" type="object[]">
  Optional list of reference images, up to 14. Omit this field for text-to-image.
  First use the [Upload Image API](../../uploads/images) to get an accessible image URL; do not pass base64 directly.

  <Expandable title="Reference image fields">
    <ParamField body="url" type="string" required>
      A publicly accessible HTTP or HTTPS image URL.
    </ParamField>
  </Expandable>

  Example: `[{"url": "https://example.com/reference-1.png"}, {"url": "https://example.com/reference-2.png"}]`.
  14 is the limit for input reference images, not the number of output images from a single request.
</ParamField>

<ParamField body="metadata" type="object">
  <Expandable title="Output resolution">
    <ParamField body="resolution" type="string" default="2K">
      Supports `1K`, `2K`, `4K`. When omitted, `2K` is used.
    </ParamField>
  </Expandable>
</ParamField>

## Request Example

```bash theme={null}
curl --request POST 'https://toapis.com/v1/images/generations' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gemini-3-pro-image-preview-vip",
    "prompt": "Keep the person from the first image, adopt the background and color palette from the second image, and generate a natural-light portrait.",
    "size": "1:1",
    "n": 1,
    "image_urls": [
      {"url": "https://example.com/reference-1.png"},
      {"url": "https://example.com/reference-2.png"}
    ],
    "metadata": {"resolution": "2K"}
  }'
```

Replace the example URLs with reference image addresses that are actually reachable. You can keep adding reference images, but the total number in the list must not exceed 14.
For a text-to-image request, simply remove `image_urls` and change the prompt.

## Querying Results

The `id` in the submit response is the task ID. Use the [Image Task Status endpoint](../../tasks/image-status) to get the status and the final image.
Task queries and [Webhook callbacks](../../webhooks/task-webhooks) follow the same conventions as the general async image endpoints.