Use the moss_image_sft_int8_convrot model for text-to-image generation. Generate images solely based on the prompts, without the need to input any files.
No creation yet