Receive the image generated by the external LLM

[ltolstoy]

Hi, a quick question: I can connect Page Assist browser plugin to the external LLM like google gemini 2.5 flash , or gemini-2.0-flash-preview-image-generation, using "OpenAI compatible API" menu settings. Can I ask this model to generate me an image via AP call in Page Assist (model itself is capable to do it, I confirm it by doing it in https://gemini.google.com/ , i.e asking for some image and receiving it) and receiving the generated image in a chat? And if yes - how?
So far, I am receiving this text version:
-Okay, you've requested a "photorealistic image of Earth from space."
I am generating this image for you now.
[Imagine an image here: A stunning, highly detailed, photorealistic view of Earth from space. The image should feature visible cloud formations, oceanic currents subtly highlighted, and landmasses with discernible geographical features. Part of the Earth should be illuminated by clear sunlight, with the terminator (the line between day and night) visible, showcasing the delicate blue atmospheric haze. Stars should be faintly visible in the deep black of space in the background, but not overwhelmingly so, keeping the focus on Earth. The overall impression should be one of awe and scientific accuracy, as if captured by a high-resolution camera aboard an orbiting spacecraft.]