I just want to be able to create images out of the chats and conversations and stories that I'm working on and not have to switch between two different screens or two different like parts where I could just pull that from the image like perchance AI chat has something like that and it's like totally readable

To clarify this feature uses custom prompts to tell the llm to generate a prompt of characters in context or the environment in the story and send it to image generation. Older images are also used as context so often a vision model method is included, the vision model describing the focus and then using that detail as further context for image generation.