Generating an image using YandexART

1. Create a service account and assign the ai.imageGeneration.user role to it.

You also need to assign the ai.languageModels.user role to the service account; in the example, we will utilize a YandexGPT Pro model to generate a prompt for YandexART.

1. Get the service account API key and save it.

   The following examples use API key authentication. Yandex Cloud ML SDK also supports IAM token and OAuth token authentication. For more information, see Authentication in Yandex Cloud ML SDK.

2. Use the pip package manager to install the ML SDK library:

   pip install yandex-cloud-ml-sdk
   

This code includes four independent examples illustrating different uses of the SDK interface:

• Example 1: A simple request consisting of a single text description.
• Example 2: A request consisting of two text descriptions, with the result saved to a file named ./image.jpeg.
• Example 3: A request consisting of two text descriptions, with weight specified.
• Example 4: A combination of a request to a YandexGPT Pro model (to generate an extended prompt) and a request to a YandexART model (to generate an image based on that prompt).

The code in the example does not return an operation object but waits for the models to execute their requests and stores the result in the result variable.

1. Create a file named generate-image.py and paste the following code into it:

   #!/usr/bin/env python3
   
   from __future__ import annotations
   import pathlib
   from yandex_cloud_ml_sdk import YCloudML
   
   message1 = "a red cat"
   message2 = "Miyazaki style"
   
   
   def main():
       sdk = YCloudML(
           folder_id="<folder_ID>",
           auth="<API_key>",
       )
   
       model = sdk.models.image_generation("yandex-art")
   
       # configuring model for all of future runs
       model = model.configure(width_ratio=1, height_ratio=2, seed=50)
   
       # Sample 1: simple run
       operation = model.run_deferred(message1)
       result = operation.wait()
       print(result)
   
       # Sample 2: run with several messages, saving the result to file
       path = pathlib.Path("./image.jpeg")
       try:
           operation = model.run_deferred([message1, message2])
           result = operation.wait()
           path.write_bytes(result.image_bytes)
       finally:
           path.unlink(missing_ok=True)
   
       # Sample 3: run with several messages specifying weight
       operation = model.run_deferred([{"text": message1, "weight": 5}, message2])
       result = operation.wait()
       print(result)
   
       # Sample 4: example of combining YandexGPT and YandexART models
       gpt = sdk.models.completions("yandexgpt")
       messages = gpt.run(
           [
               "you need to create a prompt for a yandexart model",
               "of " + message1 + "in" + message2,
           ]
       )
       print(messages)
   
       operation = model.run_deferred(messages)
       result = operation.wait()
       print(result)
   
   
   if __name__ == "__main__":
       main()
   

   Where:

   Note

   As input data for a request, Yandex Cloud ML SDK can accept a string, a dictionary, an object of the TextMessage class, or an array containing any combination of these data types. For more information, see Yandex Cloud ML SDK usage.

   • message1: Main part of the image generation prompt.
   • message2: Clarifying part of the image generation prompt.

   • <folder_ID>: ID of the folder in which the service account was created.

   • <API_key>: Service account API key you got earlier required for authentication in the API.

     The following examples use API key authentication. Yandex Cloud ML SDK also supports IAM token and OAuth token authentication. For more information, see Authentication in Yandex Cloud ML SDK.

   For more information about accessing the image generation model, see Accessing models.

2. Run the file you created:

   python3 generate-image.py
   

   Result:

   ImageGenerationModelResult(model_version='', image_bytes=<889288 bytes>)
   ImageGenerationModelResult(model_version='', image_bytes=<1062632 bytes>)
   GPTModelResult(alternatives=(Alternative(role='assistant', text='Here is an example of what a request to a YandexART model may look like:\n\n"Create an image of a red cat in Hayao Miyazaki anime style. Make the background in soft pastel shades with details emphasizing the atmosphere of magic and comfort."\n\n*Note that this is just an example and you can adapt it to your needs.*', status=<AlternativeStatus.FINAL: 3>),), usage=Usage(input_text_tokens=31, completion_tokens=76, total_tokens=107), model_version='07.03.2024')
   ImageGenerationModelResult(model_version='', image_bytes=<1180073 bytes>)