TencentTextToModelNode - ComfyUI Built-in Node Documentation

This node uses Tencent’s Hunyuan3D Pro API to generate a 3D model from a text description. It sends a request to create a generation task, polls for the result, and downloads the final model files in GLB and OBJ formats.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	The version of the Hunyuan3D model to use. The LowPoly option is unavailable for the `3.1` model.	COMBO	Yes	`"3.0"` `"3.1"`
`prompt`	The text description of the 3D model to generate. Supports up to 1024 characters.	STRING	Yes	-
`face_count`	The target number of faces for the generated 3D model. Default: 500000.	INT	Yes	3000 - 1500000
`generate_type`	The type of 3D model to generate. The available options and their associated parameters are: - Normal: Generates a standard model. Includes a `pbr` parameter (default: `False`). - LowPoly: Generates a low-polygon model. Includes `polygon_type` (`"triangle"` or `"quadrilateral"`) and `pbr` (default: `False`) parameters. - Geometry: Generates a geometry-only model.	DYNAMICCOMBO	Yes	`"Normal"` `"LowPoly"` `"Geometry"`
`seed`	A seed value for the generation. Results are non-deterministic regardless of seed. Setting a new seed controls whether the node should re-run. Default: 0.	INT	No	0 - 2147483647

Note: The generate_type parameter is dynamic. Selecting "LowPoly" will reveal additional inputs for polygon_type and pbr. Selecting "Normal" will reveal an input for pbr. Selecting "Geometry" will not reveal any additional inputs. Constraint: The "LowPoly" generate type cannot be used with the "3.1" model.

Outputs

Output Name	Description	Data Type
`model_file`	A legacy output for backward compatibility.	STRING
`GLB`	The generated 3D model in the GLB file format.	FILE3DGLB
`OBJ`	The generated 3D model in the OBJ file format.	FILE3DOBJ
`texture_image`	The texture image extracted from the generated OBJ file, if available.	IMAGE

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): 8110b6a70cdd26d5d0dcb791e654a6292895693e25076b982951c65f787422d6

​Inputs

​Outputs

Inputs

Outputs