Text component missing with gemini-3-pro-image-preview

When setting
"responseModalities": [
"Image",
"Text"
],"

This would produce a text response alongside an image on gemini-2.5.-flash-image, but does not on nano banana pro. Is this a known issue?

Is this a bug or something I’m misconfiguring?
Anyone else experiencing this?

Does anyone from Google moderate this forum, or should I be looking elsewhere for support?

I am having the same issue, it is able to produce either a text or an image but it is no longer able to output both in the same prompt. -I also tried to manipulate the chain of thought to extract text output from there but with no luck yet.

Hi @Mark_K,
Welcome to the AI Forum! Thanks for reporting this issue. We have escalated the issue to the concerned team for further investigation.