Support for responseModalities

[mucsi96]

There are models like gemini-3-pro-image-preview which support response modalities. They can return text and image.

Is there any plan to support this in Spring AI?

Example:

private static final String NANO_BANANA_PRO_MODEL = "gemini-3-pro-image-preview";

GenerateContentConfig config = GenerateContentConfig.builder()
    .responseModalities("TEXT", "IMAGE")
    .imageConfig(ImageConfig.builder()
        .aspectRatio("1:1")
        .imageSize("1K")
        .build())
    .build();

GenerateContentResponse response = googleAiClient.models.generateContent(
    NANO_BANANA_PRO_MODEL,
    prompt + ". Avoid using text.",
    config);

if (response.candidates().isEmpty() || response.candidates().get().isEmpty()) {
  throw new RuntimeException("No response from Nano Banana Pro API");
}

for (Part part : response.candidates().get().get(0).content().get().parts().get()) {
  if (part.inlineData().isPresent()) {
    return part.inlineData().get().data().get();
  }
}