Skip to main content

Overview

Generate a 3D model from a text prompt using the Rodin Gen-2.5 API. You can choose between different quality modes (Fast, Regular, or Extreme-High) to balance generation speed and output quality.

Inputs

ParameterDescriptionData TypeRequiredRange
promptText prompt describing the 3D model you want to generate.STRINGYesMax 2500 characters
modeThe generation quality and speed mode. “Fast” is quickest, “Extreme-High” produces the highest quality but takes longer.COMBOYes"Fast"
"Regular"
"Extreme-High"
materialThe material style for the generated 3D model.COMBOYes"PBR"
"Matte"
"Shiny"
geometry_file_formatThe file format for the output 3D model.COMBOYes"glb"
"obj"
"stl"
"usdz"
texture_modeTexture generation mode. “None” produces no textures, “Generated” creates standard textures, “Generated+HD” creates high-definition textures.COMBOYes"None"
"Generated"
"Generated+HD"
seedRandom seed for reproducible results. Using the same seed with the same inputs will produce the same output.INTYes0 to 2147483647
TAPoseWhether to apply T-pose (arms outstretched) to the generated model.BOOLEANYesTrue / False
hd_textureWhether to generate high-definition textures for the model.BOOLEANYesTrue / False
texture_delightWhether to apply texture delight (enhanced texture quality) to the model.BOOLEANYesTrue / False
addon_highpackWhether to generate a high-polygon version of the model in addition to the standard one.BOOLEANYesTrue / False
bbox_widthThe width of the bounding box in world units.INTYes1 to 1000
bbox_heightThe height of the bounding box in world units.INTYes1 to 1000
bbox_lengthThe length of the bounding box in world units.INTYes1 to 1000
height_cmThe height of the generated model in centimeters.INTYes1 to 300
Note: The prompt parameter must be between 1 and 2500 characters long. The seed parameter defaults to 0 (random) if not specified.

Outputs

Output NameDescriptionData Type
model_fileThe generated 3D model file in the specified format.FILE3DANY
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): 79fbaf466e9af88cdfdac0f9136a2df17ba4bc2e5bb65a35b9ad2b1181da94db