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

# Agnes Image 2.0 Flash

> A high-performance image generation and editing model for text-to-image, image-to-image, and multi-image composition workflows.

<Info>
  Agnes Image 2.0 Flash is a high-performance image generation and editing model developed by Sapiens AI. It supports text-to-image, image-to-image, and multi-image composition workflows for creative design, marketing visuals, e-commerce product images, and social content production.
</Info>

<CardGroup cols={2}>
  <Card title="Model" icon="cube">
    `agnes-image-2.0-flash`
  </Card>

  <Card title="API Endpoint" icon="link">
    `POST /v1/images/generations`
  </Card>

  <Card title="Supported Workflows" icon="wand-magic-sparkles">
    Text-to-image, image-to-image, and multi-image composition.
  </Card>

  <Card title="Current Price" icon="tag">
    Image generation is currently `$0 / image`.
  </Card>
</CardGroup>

## Overview

Agnes Image 2.0 Flash is optimized for fast, high-quality, and cost-effective image production workflows.

The model has appeared on the Artificial Analysis image editing leaderboard with an **ELO score of 1,184**, ranking in the **Top 20** range and demonstrating strong image editing capability.

## Core Capabilities

<CardGroup cols={2}>
  <Card title="Text-to-image" icon="wand-magic-sparkles">
    Generate images from text prompts.
  </Card>

  <Card title="Image-to-image" icon="images">
    Edit, transform, or enhance existing images.
  </Card>

  <Card title="Multi-image Input" icon="layer-group">
    Use multiple reference images to create a new image.
  </Card>

  <Card title="Image Editing" icon="paintbrush">
    Modify composition, style, objects, backgrounds, and scenes.
  </Card>

  <Card title="Style Control" icon="palette">
    Control artistic style, lighting, layout, and visual direction.
  </Card>

  <Card title="Fast Generation" icon="bolt">
    Designed for frequent, fast, production-grade creative workflows.
  </Card>
</CardGroup>

## Use Cases

<CardGroup cols={2}>
  <Card title="Creative Design" icon="pen-ruler">
    Posters, concept art, and social media visuals.
  </Card>

  <Card title="Marketing Content" icon="bullhorn">
    Product ads, campaign creatives, and banners.
  </Card>

  <Card title="Image Editing" icon="paintbrush">
    Object replacement, background replacement, style transfer, and local edits.
  </Card>

  <Card title="Character Composition" icon="users">
    Combine multiple characters or reference images into one scene.
  </Card>

  <Card title="E-commerce" icon="cart-shopping">
    Product image enhancement, product staging, and marketing hero images.
  </Card>

  <Card title="Social Content" icon="share-nodes">
    Memes, avatars, thumbnails, and lifestyle visual assets.
  </Card>
</CardGroup>

## API Reference

### Endpoint

```text theme={null}
POST https://api.agnes-ai.cn/v1/images/generations
```

### Headers

```bash theme={null}
-H "Authorization: Bearer YOUR_API_KEY"
-H "Content-Type: application/json"
```

### Request Parameters

| Parameter                    | Type      | Required                    | Description                                                        |
| ---------------------------- | --------- | --------------------------- | ------------------------------------------------------------------ |
| `model`                      | string    | Yes                         | Model name. Use `agnes-image-2.0-flash`.                           |
| `prompt`                     | string    | Yes                         | Text prompt describing the target image or editing instruction.    |
| `size`                       | string    | Yes                         | Output image size, such as `1024x768`, `1024x1024`, or `768x1024`. |
| `image`                      | string\[] | Required for image-to-image | Input image array. Supports public URLs or Data URI Base64.        |
| `return_base64`              | boolean   | No                          | Use this when text-to-image output should be returned as Base64.   |
| `extra_body.response_format` | string    | No                          | Output format. Common values are `url` and `b64_json`.             |

## Important Notes

<Warning>
  Do not place `response_format` at the top level of the request body. For URL or Base64 output, place it inside `extra_body`.
</Warning>

<CardGroup cols={2}>
  <Card title="Text-to-image" icon="wand-magic-sparkles">
    Only `model`, `prompt`, and `size` are required. Do not pass `image`.
  </Card>

  <Card title="Image-to-image" icon="images">
    Pass image URLs or Data URI Base64 values through `extra_body.image`.
  </Card>

  <Card title="No tags Required" icon="ban">
    Image-to-image requests do not require `tags: ["img2img"]`.
  </Card>

  <Card title="Safe Examples" icon="key">
    Use `YOUR_API_KEY` in public documentation. Do not expose real API keys.
  </Card>
</CardGroup>

## Request Examples

<Tabs>
  <Tab title="Text-to-image: URL Output">
    ```bash theme={null}
    curl https://api.agnes-ai.cn/v1/images/generations \
      -H "Authorization: Bearer YOUR_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "agnes-image-2.0-flash",
        "prompt": "A clean product photo of a glass cube on a white studio background, soft shadows, high detail",
        "size": "1024x768",
        "extra_body": {
          "response_format": "url"
        }
      }'
    ```
  </Tab>

  <Tab title="Text-to-image: Base64 Output">
    ```bash theme={null}
    curl https://api.agnes-ai.cn/v1/images/generations \
      -H "Authorization: Bearer YOUR_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "agnes-image-2.0-flash",
        "prompt": "A clean product photo of a glass cube on a white studio background, soft shadows, high detail",
        "size": "1024x768",
        "return_base64": true
      }'
    ```
  </Tab>

  <Tab title="Image-to-image: URL Input + URL Output">
    ```bash theme={null}
    curl https://api.agnes-ai.cn/v1/images/generations \
      -H "Authorization: Bearer YOUR_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "agnes-image-2.0-flash",
        "prompt": "Transform this image into a cinematic cyberpunk style while preserving the main subject and composition",
        "size": "1024x768",
        "extra_body": {
          "image": [
            "https://example.com/input-image.png"
          ],
          "response_format": "url"
        }
      }'
    ```
  </Tab>

  <Tab title="Image-to-image: Base64 Output">
    ```bash theme={null}
    curl https://api.agnes-ai.cn/v1/images/generations \
      -H "Authorization: Bearer YOUR_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "agnes-image-2.0-flash",
        "prompt": "Make the object orange while preserving the original composition",
        "size": "1024x768",
        "extra_body": {
          "image": [
            "https://example.com/input.png"
          ],
          "response_format": "b64_json"
        }
      }'
    ```
  </Tab>

  <Tab title="Multi-image Composition">
    ```bash theme={null}
    curl https://api.agnes-ai.cn/v1/images/generations \
      -H "Authorization: Bearer YOUR_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "agnes-image-2.0-flash",
        "prompt": "Combine the two characters into an intense fantasy battle scene, dynamic lighting, detailed background, cinematic composition",
        "size": "1024x768",
        "extra_body": {
          "image": [
            "https://example.com/character-1.png",
            "https://example.com/character-2.png"
          ],
          "response_format": "url"
        }
      }'
    ```
  </Tab>
</Tabs>

## Response Format

<Tabs>
  <Tab title="URL Output">
    ```json theme={null}
    {
      "created": 1780000000,
      "data": [
        {
          "url": "https://storage.googleapis.com/agnes-aigc/xxx.png",
          "b64_json": null,
          "revised_prompt": null
        }
      ]
    }
    ```
  </Tab>

  <Tab title="Base64 Output">
    ```json theme={null}
    {
      "created": 1780000000,
      "data": [
        {
          "url": null,
          "b64_json": "iVBORw0KGgoAAAANSUhEUgAA...",
          "revised_prompt": null
        }
      ]
    }
    ```
  </Tab>
</Tabs>

### Response Fields

| Field                   | Type          | Description                                                   |
| ----------------------- | ------------- | ------------------------------------------------------------- |
| `created`               | integer       | Request creation timestamp.                                   |
| `data`                  | array         | List of generated image results.                              |
| `data[].url`            | string / null | Generated image URL. Usually `null` when using Base64 output. |
| `data[].b64_json`       | string / null | Base64 image data. Usually `null` when using URL output.      |
| `data[].revised_prompt` | string / null | Revised prompt if available. Otherwise `null`.                |

## Best Practices

<AccordionGroup>
  <Accordion title="Text-to-image Prompts">
    Provide clear visual instructions, including subject, scene, style, lighting, composition, and quality requirements.

    ```text theme={null}
    [Subject] + [Scene / Background] + [Style] + [Lighting] + [Composition] + [Quality Requirements]
    ```

    ```text theme={null}
    A professional product photo of a wireless headphone on a clean white background, soft studio lighting, sharp details, commercial photography style
    ```
  </Accordion>

  <Accordion title="Image Editing Prompts">
    Clearly describe what should change and what should remain unchanged.

    ```text theme={null}
    [Editing Instruction] + [Elements to Preserve] + [Target Style / Scene] + [Lighting] + [Composition] + [Quality Requirements]
    ```

    ```text theme={null}
    Change the background to a futuristic city at night while keeping the person's face, outfit, and pose unchanged
    ```
  </Accordion>

  <Accordion title="Multi-image Composition Prompts">
    Describe the relationship between the input images and how they should be combined.

    ```text theme={null}
    Place the person from the first image beside the robot from the second image in a cinematic sci-fi battle scene
    ```
  </Accordion>
</AccordionGroup>

## FAQ

<AccordionGroup>
  <Accordion title="Does Agnes Image 2.0 Flash support text-to-image?">
    Yes. Text-to-image requests only require `model`, `prompt`, and `size`. The `image` parameter is not needed.
  </Accordion>

  <Accordion title="Does Agnes Image 2.0 Flash support image-to-image?">
    Yes. Image-to-image requests require input images in `extra_body.image`.
  </Accordion>

  <Accordion title="Do image-to-image requests require tags?">
    No. Do not pass `tags: ["img2img"]`. Use `model`, `prompt`, `size`, and `extra_body.image` instead.
  </Accordion>

  <Accordion title="Why does top-level response_format cause an error?">
    In this API structure, `response_format` should be placed inside `extra_body`. Placing it at the top level may cause a `400` error.
  </Accordion>

  <Accordion title="What if the input image URL cannot be accessed?">
    Use a publicly accessible HTTPS image URL. If the image cannot be made public, use Data URI Base64 input.
  </Accordion>

  <Accordion title="How should client timeouts be configured?">
    Image generation may take several seconds to tens of seconds. A client timeout of `60s - 360s` is recommended.
  </Accordion>
</AccordionGroup>

## Pricing

| Type             | Standard Price   | Current Price |
| ---------------- | ---------------- | ------------- |
| Image generation | `$0.003 / image` | `$0 / image`  |

## Integration Checklist

<Check>
  Use `agnes-image-2.0-flash` as the model name.
</Check>

<Check>
  Use `https://api.agnes-ai.cn/v1/images/generations` as the API endpoint.
</Check>

<Check>
  For text-to-image, include `model`, `prompt`, and `size`.
</Check>

<Check>
  For image-to-image, provide input images through `extra_body.image`.
</Check>

<Check>
  Do not place `response_format` at the top level of the request body.
</Check>