HiDream O1 Image

HiDream O1 Image is an open image generation foundation model built on a Pixel-level Unified Transformer that encodes raw pixels, text, and task conditions in a single token space rather than using a separate VAE or text encoder. It handles text-to-image, image editing, and subject-driven personalization at resolutions up to 2048 by 2048, with a focus on accurate long-text rendering and identity preservation across new scenes.