A lot to unwrap here 
Gemini will eventually ship a text to image modality. When that happens we will be able to ask it to generate images.
Gemini models all have the image modality for input, so as long as you enable vision on bot and llm it will understand images you paste.
You can then use either Flux or Dall E or stable diffusion to generate images in chat.
So … yes … it can draw a picture via a tool