Generate images using Bytedance's SeedDream 3.0 model. Native 2K high resolution output with exceptional text layout, accurate text generation, photorealistic portraits, and fast generation (3 seconds for 1K images).
Approx. Price
$0.040 per image
Model Type
text-to-image
Preview Examples
3
Generation controls available for this model.
Images Per Run
Up to 8
Resolution Options
8
1024x1024 (Square (1:1)), 1152x896 (Landscape (4:3)), 896x1152 (Portrait (3:4)) +5 more
Tunable Settings
5
Enable Safety Checker
Default
Yes
Guidance Scale
Default
2.5
Controls how closely the output image aligns with the input prompt. Higher values mean stronger prompt correlation.
Number of Images
Default
1
Resolution
Default
1024x1024
Options (8)
1024x1024 (Square (1:1)), 1152x896 (Landscape (4:3)), 896x1152 (Portrait (3:4)), 1344x768 (Landscape (16:9)) +4 more
Seed
Default
N/A
Random seed to control the stochasticity of image generation.
Human preference benchmarks sourced from Artificial Analysis.
Text to Image
#23 / 129
ELO
1155.0
Appearances
4,923
95% CI
-9/9
Release Date 2025-04 · Matched as Seedream 3.0
Artificial Analysis API
Fisheye lens, the head of a cat, the image shows the effect that the facial features of the cat are distorted due to the shooting method.
Default

A man sits on a grassy hill at dusk, watching a sprawling city skyline light up below. The grass is sharp in the foreground, the city distant but glowing. Photorealistic, entle cinematic color grading."
Default

digital painting of a marble corridor framed by three arched portals. left portal: a desert with a dinosaur skeleton half-buried in dunes; center: trees with green-violet auroras swirling above; right: coral reefs with jellyfish
Default