Python: feature: support gpt-image-1

[ymuichiro]

This pull request was created in response to the issue #12500 (comment). The current AzureTextToImage implementation only works with DALL-E 3. For gpt-image-1, the response format has changed from a URL to base64 only, which the current code does not support.

Additionally, gpt-image-1 introduces a new image editing feature that also needs to be supported. Since breaking changes are required, new methods have been added.

Minimal code that reproduces the problem

service = AzureTextToImage(
        service_id=service_id,
        deployment_name="image-1",
        endpoint=AZURE_OPENAI_IMAGE_ENDPOINT,
        api_key=AZURE_OPENAI_IMAGE_API_KEY,
    )
)

settings = service.get_prompt_execution_settings_class()(service_id="image1")
settings.prompt = "sky"
settings.size = ImageSize(width=1024, height=1024)
settings.quality = "low"
r = await service.generate_image(settings=settings)

Example of use with newly added methods

from semantic_kernel.connectors.ai.open_ai import AzureTextToImage
service = AzureTextToImage(
    service_id="image1",
    deployment_name="gpt-image-1",
    endpoint=AZURE_OPENAI_IMAGE_ENDPOINT,
    api_key=AZURE_OPENAI_IMAGE_API_KEY,
    api_version="2025-04-01-preview",
)
settings = service.get_prompt_execution_settings_class()(service_id="image1")
settings.n = 3
images_b64 = await service.generate_images("A cute cat wearing a whimsical striped hat", settings=settings)

  from semantic_kernel.connectors.ai.open_ai import AzureTextToImage
  service = AzureTextToImage(
      service_id="image1",
      deployment_name="gpt-image-1",
      endpoint=AZURE_OPENAI_IMAGE_ENDPOINT,
      api_key=AZURE_OPENAI_IMAGE_API_KEY,
      api_version="2025-04-01-preview",
  )
  file_paths = ["./new_images/img_1.png", "./new_images/img_2.png"]
  settings = service.get_prompt_execution_settings_class()(service_id="image1")
  settings.n = 2
  results = await service.edit_image(
      prompt="Make the cat wear a wizard hat",
      image_paths=file_paths,
      settings=settings,
  )

Problems Identified

1. Assumption of URL-based responses. The current implementation assumes a response format that includes an image url, which is not the case for gpt-image-1. See:

   semantic-kernel/python/semantic_kernel/connectors/ai/open_ai/services/open_ai_text_to_image_base.py

   Line 69 in 8d1b3fd

   raise ServiceResponseException("Failed to generate image.")

[moonbox3]

Thanks for working on this, @ymuichiro. Are there unit tests we can add so we have coverage for the new code?

[ymuichiro]

@moonbox3

I've added unit tests!
ymuichiro@5e0269d

I also made some minor adjustments to other parts of the code as I found issues in existing test code.

[ymuichiro]

@moonbox3

The test was failing, so I fixed it.

1. An error was occurring on the AzureTextToImage side. This has been resolved.
2. responses.usage is assumed to always return True for hasattr, but a safety check has been added just in case.

[markwallace-microsoft]

Python Test Coverage Report •

File	Stmts	Miss	Cover	Missing
connectors/ai/open_ai/prompt_execution_settings
open_ai_text_to_image_execution_settings.py	45	2	95%	54, 63
connectors/ai/open_ai/services
open_ai_handler.py	120	19	84%	150–151, 156–159, 164, 172–173, 189–190, 202, 211–212, 225–229
open_ai_text_to_image_base.py	101	15	85%	47, 51, 55, 63, 112, 114, 119, 128, 136–137, 139, 142, 206, 240, 244
TOTAL	26507	4529	82%

Python Unit Test Overview

Tests	Skipped	Failures	Errors	Time
3649	22 💤	0 ❌	0 🔥	1m 53s ⏱️