ByteDance Seedream 5

A ComfyUI node in Seedance/Image Generation with 14 inputs and 3 outputs.

By theclueless-ai·Created 4 months ago·Updated 4 months ago· 0

◄api_key►

◄promptVibrant close-up editorial portrait, sharp focus on eyes, dramatic studio lighting.►

◄modelseedream-5-0-260128►

◄size_preset2K 1:1►

◄width2048►

◄height2048►

◄sequential_image_generationfalse►

◄max_images1►

◄seed0►

◄watermarkfalse►

◄fail_on_partialfalse►

◄output_formatpng►

◄image_url►

CategorySeedance/Image Generation

Inputs (14)

Name	Type	Default	Description
api_key	STRING		BytePlus ARK API key. Leave empty to use ARK_API_KEY env variable.
prompt	STRING	Vibrant close-up editorial portrait, sharp focus on eyes, dramatic studio lighting.	Text description of the image to generate.
model	COMBO	seedream-5-0-260128	seedream-5-0-260128 (full) or seedream-5-0-lite-260128 (lite).
size_preset	COMBO	2K 1:1	Aspect-ratio + resolution preset. Choose 'custom' to use the width/height inputs instead.
width	INT	2048512–4704	Custom image width (only used when size_preset = 'custom').
height	INT	2048512–4704	Custom image height (only used when size_preset = 'custom').
sequential_image_generation	BOOLEAN	false	Enable grouped image generation. When True the model can return up to max_images related images in a single request.
max_images	INT	11–15	Maximum generated images (only used when sequential_image_generation is True). Reference images + generated images must be ≤ 15.
seed	INT	00–18446744073709550000	Random seed. 0 = server picks a random seed.
watermark	BOOLEAN	false	Add a BytePlus watermark to the output image.
fail_on_partial	BOOLEAN	false	When sequential generation returns fewer images than requested, raise an error instead of returning the partial batch.
output_format	COMBO	png	2 options: png, jpeg
imageopt	IMAGE		Optional reference image(s). A batched IMAGE is sent as multiple reference images (multi-reference mode).
image_urlopt	STRING		Optional HTTP URL(s) to use instead of the IMAGE input. Separate multiple URLs with newlines.