Hunyuan3Dv2ConditioningMultiView - ComfyUI Built-in Node Documentation

The Hunyuan3Dv2ConditioningMultiView node combines CLIP vision outputs from up to four views (front, left, back, and right) into a single multi-view conditioning.

The Hunyuan3Dv2ConditioningMultiView node combines CLIP vision outputs from up to four views (front, left, back, and right) into a single multi-view conditioning. Each provided view has a positional encoding added to its CLIP vision embedding, and the resulting embeddings are concatenated. The node outputs a positive conditioning based on the combined embeddings and a negative conditioning filled with zeros of the same shape.

Inputs

Parameter	Description	Data Type	Required	Range
front	CLIP vision output for the front view. Optional view input.	CLIP_VISION_OUTPUT	No	-
left	CLIP vision output for the left view. Optional view input.	CLIP_VISION_OUTPUT	No	-
back	CLIP vision output for the back view. Optional view input.	CLIP_VISION_OUTPUT	No	-
right	CLIP vision output for the right view. Optional view input.	CLIP_VISION_OUTPUT	No	-

Note: At least one view input must be provided for the node to function. The node only processes views that contain valid CLIP vision output data and skips views that are not connected. Each view receives a fixed positional encoding based on its slot (front, left, back, right), and the processed embeddings from all provided views are joined together along the sequence dimension.

Outputs

Output Name	Description	Data Type
positive	Positive conditioning containing the combined multi-view embeddings with positional encoding.	CONDITIONING
negative	Negative conditioning with zero values matching the positive conditioning shape.	CONDITIONING

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): 1492b51661d0bb8f2c142c1b1e8ef104beed1b9dae532a970e2928e27ad71d69