About
Seedream is a family of text-to-image models developed by the Seed team at ByteDance. It spans successive generations — Seedream 2.0, Seedream 3.0 and Seedream 4.0 — with native bilingual capability (Chinese and English, including text rendering in images), high-resolution generation (2K/4K) and — from version 4.0 — unified image generation and editing. The models are available in the Doubao and Dreamina apps and via the Volcano Engine API, among others.
