Image generation
Available now

Seedream 5.0 Lite

by ByteDance

ByteDance's image model that reads the prompt carefully and pulls in live web context, for news‑aware and topical visuals.

Seedream 5.0 Lite

by ByteDance

Key features

Technical specifications

5.0 Lite

Lite variant of Seedream 5.0

Deep thinking

Plans before generating

Online search

Live web information at generation time

Prompt-following

Tested on ByteDance's MagicBench

Use cases

Topical and news-aware visuals

Generate images tied to current events, fresh launches, or trending references. Online search context bypasses the training cutoff.

Reasoning-heavy prompts

Prompts that need careful reading before drawing. Multi-element scenes, layered instructions, or anything where older models drop detail.

Typography and UI design

Generate posters, UI mockups, and chart-style visuals where typography and structured layout need to render coherently inside the image.

Editorial and photography

Standard editorial and photography prompts benefit from the same deep thinking. Each prompt is read carefully before pixels are committed.

Prompt examples

Topical visual

Editorial illustration of a recently launched flagship device on a designer desk, top-down flat lay, accurate to current real-world references

Edit prompt

Reasoning-heavy

A bookshop window in autumn, hand-lettered chalkboard sign with the word "new arrivals" in clean script, warm tungsten light, photoreal

Edit prompt

Typography-led

Vintage travel poster for Lisbon, large condensed title across the top, small descriptive copy along the bottom, warm muted palette

Edit prompt

Overview

Seedream 5.0 Lite is ByteDance's unified multimodal image generation model, and the most distinct shift in the Seedream line. Two capabilities define it: deep-thinking reasoning before generation, the model interprets the prompt carefully before committing to an image, and built-in online search, so prompts tied to current events or fresh references can be visualized with live web context rather than being limited to training-cutoff knowledge. On Morphic, both behaviors live inside a single model selection in Copilot, no separate research or context-injection step needed.

Simple pricing

Get started for free today, with the option to upgrade or cancel anytime.

Basic

$9/ month
billed as $0 per year

1100 monthly credits

1 user only

All models

Workflows

Standard

$24/ month
billed as $0 per year

3625 monthly credits

1 user only

All models

Workflows

Pro

$45/ month
billed as $0 per year

6350 shared monthly credits

1 user

+ up to 4 more at extra cost

All models

Workflows

Pro Max

$170/ month
billed as $0 per year

24650 shared monthly credits

1 user

+ up to 9 more at extra cost

All models

Workflows

Enterprise

For higher limits

Custom

pricing and billing terms

High-volume credits
Custom seat limits
All models
Workflows
Pricing Gradient

Free

For playing around

$0

forever free

Up to 20 credits
1 user only
Limited models
Workflows
Compare plan details

FAQs

What is Seedream 5.0 Lite?
Seedream 5.0 Lite is ByteDance's unified multimodal image generation model with two distinguishing capabilities, deep-thinking reasoning before generation, and built-in online search for live web context at generation time.
What does online search do in Seedream 5.0 Lite?
The model can pull in real-time information, current events, trending topics, and recently released references, and visualize them. Useful when a prompt depends on fresh context that wouldn't be in a typical model's training data.
How is Seedream 5.0 Lite different from Seedream 4.5?
4.5 is a fidelity-focused upgrade with multi-image references and dense text rendering. 5.0 Lite shifts the focus to reasoning and live context, pick 5.0 Lite when the prompt needs careful interpretation or pulls on current real-world information.
How do I use Seedream 5.0 Lite on Morphic?
Open Copilot, describe the image, and select Seedream 5.0 Lite. Reach for it when a prompt benefits from careful reasoning, structured typography, or live web context, otherwise Seedream 4.5 is the faster default for general image generation.
Is there a non-Lite Seedream 5.0?
ByteDance has publicly released the Lite variant. The Lite tag positions this as the accessible model in the 5.0 line, capabilities like deep thinking and online search are present in the released model.
Does Seedream 5.0 Lite pair with Seedance for video?
Yes. A typical workflow is generating a context-aware still in Seedream 5.0 Lite, then carrying it into Seedance 1.5 Pro or Seedance 2.0 for image-to-video on Morphic, both lines share ByteDance lineage.

Other models

Explore the rest of the Morphic model catalog.

See all models →

ChatGPT Images 2.5

OpenAI

OpenAI's image model, released September 2026. Sharper detail, precise edits, faster.

Lyria 3.5

Google DeepMind

Google's best-sounding music model. Full songs with structure, vocals, and lyrics.

MiniMax H3 Max Turbo

fal.ai

The fast tier of fal's post-trained H3. Twice the speed, nearly the same look.

Inworld TTS 2

Inworld

Ninety-five voices, 100+ languages. Speech that starts in under a second.