ByteDance's unified multimodal image model for creating images from text and intelligent photo editing. Stronger prompt understanding, consistent subjects, and controllable visuals — from marketing shots to knowledge diagrams.
Start GeneratingSeedream 5.0 Lite combines generation and editing with shared multimodal reasoning. Create from a text prompt in eight aspect ratios, or upload up to 14 reference images and describe how you want them changed while preserving structure.
Built for visual reasoning and deep understanding — not just pixel matching. It interprets vague briefs, respects real-world logic, and keeps edits localized to what you ask for. Popular for marketing creatives, education graphics, product mockups, and editorial illustrations.
Choose create or edit, set aspect ratio and quality (Basic 2K or High 4K), sign in, and generate for 11 credits per image. Processing typically completes in under a minute. Download results for personal or commercial projects under our terms.
Try creating from text and editing photos with sample prompts, then scale your workflow at 11 credits per image.
