ByteDance releases Dreamina 4.0, its first version to support multimodal image generation.

ByteDance’s Dreamina AI has released Dreamina Image 4.0, the first version to support multimodal image generation. With a single model, it enables text-to-image creation, image editing, and multi-image generation. The text-to-image feature offers stronger instruction-following, higher resolution, and faster generation speed; image editing allows modifications to single or multiple images through natural language commands; while multi-image generation can produce a set of related images in one go.