Want an ad-free experience? Upgrade your plan.
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.
Fine-tuned for dreamy, cyberpunk, and anime fantasy wallpaper scenes.
Fine-tuned for dreamy, cyberpunk, and anime fantasy wallpaper scenes.
Basic and High both cost 11 credits per image.
Your generated image will appear here
Seedream 5.0 Lite combines generation and editing with shared multimodal reasoning
Describe your scene in 3–3,000 characters and generate photorealistic or stylized images in eight aspect ratios. Ideal for ads, product heroes, portraits, and concept art with strong instruction following.
Upload up to 14 reference images and instruct the model to change materials, lighting, style, or scene context while preserving structure. Perfect for iterative design and controlled variations.
Choose Basic for efficient 2K outputs or High for sharper 4K results. Both modes use the same flat 11-credit pricing on our platform — pick quality for resolution, not cost
Seedream 5.0 Lite is built for multimodal reasoning and deep visual understanding — not just pixel matching. It interprets vague briefs, respects real-world logic, and keeps edits localized to what you ask for.
Multi-step understanding of composition, spatial relationships, and prompt intent — fewer retries and more coherent frames.
Better alignment with real-world structure for diagrams, educational graphics, and professional layouts that need to look plausible.
Apply reference aesthetics — lighting, mood, and artistic tone — without writing overly technical prompts.
Short edit instructions are expanded like a designer would: change only what you specify and keep the rest intact.
More stable generation when several subjects or detailed spatial relationships appear in one prompt.
Turn complex ideas into clearer diagrams, slides, and explainer-style graphics for education and business.
Three steps from prompt to download
Generate a new image from a text prompt, or upload reference images (up to 14) and describe how you want them changed.
Select 1:1 through 21:9, then Basic (2K) or High (4K). Use sample prompts and images to test quickly.
Sign in, click Generate, and download your result when processing completes — typically under a minute.
Two live generations from Van Gogh Studio — create from text and edit existing photos — not stock images.
Create from Text
Prompt
Cinematic alpine landscape at golden hour, dramatic clouds, ultra-detailed, 21:9 widescreen.
21:9 aspect ratio · Basic (2K) · 11 credits
Edit Images
Prompt
Place the product on a warm wooden counter in a sunlit modern kitchen. Add soft morning light and gentle steam in the background. Keep the label sharp and readable.
1:1 aspect ratio · Basic (2K) · 11 credits
Seedream 5.0 Lite on Van Gogh Studio
Ready to try Seedream 5.0 Lite?
Try creating from text and editing photos with sample prompts, then scale your workflow at 11 credits per image.
Free AI tools to go from idea to publish — pick a tool and start creating in minutes.
Upload a still and watch AI bring it to life with natural motion.
Turn still images into cinematic video
Open
Set two key frames — AI fills the smooth transition between them.
Turn still images into cinematic video
Open
Fine-tune images with AI editing
Open
Fix blur, noise & detail in one click — old photos look brand-new.
Enhance image quality before publishing
Open
Create product shots for e-commerce
Open
Browse prompt templates and ideas
OpenExplore