Alibaba Wan 2.7 Text To Image (Official Edition)

A ComfyUI node in BizyAirModelServices/BizyTRD/Wan Image with 10 inputs and 3 outputs.

By siliconflow·Created 4 months ago·Updated 19 days ago· 20

◄prompt►

◄size2K►

◄color_palette►

◄seed0►

◄custom_width2048►

◄custom_height2048►

◄watermarkfalse►

◄thinking_modetrue►

◄skip_errorfalse►

CategoryBizyAirModelServices/BizyTRD/Wan Image

Inputs (10)

Name	Type	Default	Description
prompt	STRING		Text prompts. Support for Chinese and English, no more than 5000 characters. Describe the content of the screen you want to generate.
size	COMBO	2K	Output picture resolution. 1K = 1024 × 1024,2K = 2048 × 2048,4K = 4096 × 4096(Pro) . 4K is not supported in group mode.
color_paletteopt	STRING		Custom color theme JSON array, containing 3-10 colors, 8 recommended. Each color contains the hex and ratio fields. The sum of all ratios must be 100.00% . Available only when group mode is turned off.
seedopt	INT	0-1–2147483647	Random number seed, value range 0-2147483647, -1 means automatic generation. The same seed keeps the generated content relatively stable.
custom_widthopt	INT	2048768–4096	Take effect only if size = Custom. Total pixels and aspect ratio must be within limits.
custom_heightopt	INT	2048768–4096	Take effect only if size = Custom. Total pixels and aspect ratio must be within limits.
watermarkopt	BOOLEAN	false	Whether to add“AI generated” watermark logo, watermark is located in the lower right corner of the picture.
thinking_modeopt	BOOLEAN	true	Turn on the model to enhance the reasoning ability to improve the quality of the map, but will increase the time-consuming. This only takes effect when group mode is turned off.
api_configopt	BIZYAIR_OPENAPI_CONFIG		单次请求覆盖base_url和api_key
skip_erroropt	BOOLEAN	false	开启后遇到错误不中断工作流，输出对应类型的错误占位符