Seedream 4.0 is ByteDance's unified image creation model released in September 2025, the first in the Seedream series to integrate text-to-image generation and image editing into a single architecture. Built on a Mixture-of-Experts Diffusion Transformer backbone, it generates images at up to 4K resolution (2K default) in approximately 1.8 seconds — over ten times faster than its predecessor. The model supports up to six reference images, produces up to nine consistent outputs per batch, and excels at complex typographic rendering in English and Chinese across more than thirty artistic styles.
SeedreamSeedream 4.0 is ByteDance's unified image creation model released in September 2025, the first in the Seedream series to integrate text-to-image generation and image editing into a single architecture. Built on a Mixture-of-Experts Diffusion Transformer backbone, it generates images at up to 4K resolution (2K default) in approximately 1.8 seconds — over ten times faster than its predecessor. The model supports up to six reference images, produces up to nine consistent outputs per batch, and excels at complex typographic rendering in English and Chinese across more than thirty artistic styles.
Seedream