Seedream 5.0 Lite

Reasoning-first image model with web search; generation grounded in current information.

modellane/seedream-5.0-lite · Text to image · ByteDance

ByteDance's reasoning-first image model. It produces images grounded in current information through real-time web search and chain-of-thought, and keeps place, physics, and light consistent. Built-in editing support lets you do style transfer and multi-source composition on the same model.

Vendor
ByteDance
Task
Text to image
Output
Image
Price
$0.04 · per image

Input parameters

Recommended use cases

Where it is strong

Limits

Example prompt

The interior of a modern Turkish coffee shop in Kadıköy, large windows, morning light, people working on laptops, barista pouring latte art, the words "KADIKÖY" on the window in clean typography, editorial photography

Example request

POST /v1/queue/modellane/seedream-5.0-lite

Playground

The identifier, price, parameters, and example requests on this page can be read without signing in. Running the model from the panel sends a real generation request and charges the project balance; the playground therefore requires an account.