> ## Documentation Index
> Fetch the complete documentation index at: https://docs.g4f.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Image Generation

> Learn how to generate images using the G4F library and various image generation providers.

<Card title="Image Generation" icon="image">
  Generate stunning images using the power of AI with G4F's image generation capabilities.
</Card>

## Introduction

G4F provides a simple and intuitive way to generate images using different image generation providers. With just a few lines of code, you can create visually appealing images based on text prompts.

<Info>
  To unleash the full potential of image generation, ensure you have installed the necessary dependencies by running:

  ```bash theme={null}
  pip install g4f[image]
  ```
</Info>

## Supported Providers

G4F supports the following image generation providers:

<CardGroup cols={2}>
  <Card title="Bing" icon="microsoft">
    Utilizes Microsoft's Image Creator for image generation.
  </Card>

  <Card title="Google Gemini" icon="google">
    Available for free accounts with IP addresses outside Europe.
  </Card>

  <Card title="OpenAI Chat with GPT-4" icon="robot">
    Accessible for users with a Plus subscription.
  </Card>
</CardGroup>

## Generating Images

To generate images using G4F, follow these steps:

<Steps>
  <Step title="Import the necessary modules" stepNumber={1}>
    ```jsx theme={null}
    from g4f.client import Client
    from g4f.Provider import BingCreateImages, OpenaiChat, Gemini

    client = Client(
        image_provider=Gemini,
        api_key='api-key if needed'
        ...
    )
    ```

    <Info>
      When using the `OpenaiChat` provider, you need to [**provide an access token**](/docs/core/usage/auth_cookies) from a Plus user.
    </Info>
  </Step>

  <Step title="Set up the request for image creation" stepNumber={2}>
    ```jsx theme={null}
    response = client.images.generate(
        model="dall-e-3",
        prompt="a white siamese cat",
        ...
    )
    ```

    ***

    <Card title="Request Parameters" icon="sliders">
      <ParamField path="model" type="str" default="g4f.models.default">
        The model to use for image generation. Defaults to the default model.
      </ParamField>

      <ParamField path="prompt" type="text">
        The text prompt to use for image generation.
      </ParamField>

      <ParamField path="kwargs" type="json">
        Additional keyword arguments to pass to the image generation provider.
      </ParamField>
    </Card>
  </Step>

  <Step title="Get image links from the response" stepNumber={3}>
    ```jsx theme={null}
    image_url = response.data[0].url
    print(image_url) 
    ```
  </Step>

  <Step title="Image Variations" stepNumber={4}>
    ```jsx theme={null}
    response = client.images.create_variation(
        image=open("cat.jpg", "rb"),
        model="bing",
        ...
    )

    image_url = response.data[0].url
    ```

    ***

    <Card title="Request Parameters" icon="sliders">
      <ParamField path="image" type="file">
        The image to use for creating variations.
      </ParamField>

      <ParamField path="model" type="model" default="ImageType">
        The model to use for image generation. Defaults to the default model.
      </ParamField>

      <ParamField path="kwargs" type="json">
        Additional keyword arguments to pass to the image generation provider.
      </ParamField>
    </Card>
  </Step>
</Steps>

<Tip>
  Experiment with different providers and prompts to generate a variety of images!
</Tip>

## Example

Here's an example of generating an image using the Bing provider:

```jsx theme={null}
from g4f.client import Client

client = Client()

response = client.images.generate(
    model="dall-e-3",
    prompt="a white siamese cat",
    ...
)

print(response.data[0].url)
```

<Frame>
  <img src="https://raw.githubusercontent.com/xtekky/gpt4free/main/docs/cat.jpeg" alt="Generated Image" />

  <img src="https://raw.githubusercontent.com/xtekky/gpt4free/main/docs/cat.webp" />
</Frame>

## Chat with images

<Steps>
  <Step title="Import the necessary modules" stepNumber={1}>
    ```jsx theme={null}
    from g4f.client import Client
    from g4f.Provider.GeminiPro import GeminiPro

    client = Client(
      api_key="...",
      provider=GeminiPro
    )
    ```

    <Info>
      When using the `GeminiPro` provider, you need to [**provide an api-key**](/docs/core/usage/auth_cookies).
    </Info>
  </Step>

  <Step title="Chat with your image" stepNumber={1}>
    ```jsx theme={null}
    response = client.chat.completions.create(
        model="gemini-pro-vision",
        messages=[{"role": "user", "content": "What can you see on this image"}],
        image=open("image.png", "rb"))

    print(response.choices[0].message.content)
    ```

    ```bash theme={null}
    user@mac-air:~$ python3 main.py
    >>> I can see a bird flying in the tropical forest
    ```
  </Step>
</Steps>

## Conclusion

G4F makes it easy to generate images using various providers. With just a few lines of code, you can create stunning visuals based on text prompts. Explore the different providers and unleash your creativity!

<Tip>
  Remember to handle the generated images appropriately and respect copyright and usage rights.
</Tip>