TripoSplatConditioning - ComfyUI Built-in Node Documentation

Complete documentation for the TripoSplatConditioning node in ComfyUI. Learn its inputs, outputs, parameters and usage.

This node encodes an input image using DINOv3 and the Flux2 VAE to create positive and negative conditioning data for the TripoSplat model. It also generates a fixed-size noise target (latent plus camera data) that serves as the starting point for the KSampler.

Inputs

Parameter	Description	Data Type	Required	Range
clip_vision	DINOv3 ViT-H/16+ image encoder	CLIP_VISION	Yes	-
vae	Flux2 VAE	VAE	Yes	-
image	The input image to encode	IMAGE	Yes	-

Outputs

Output Name	Description	Data Type
positive	Positive conditioning data containing DINOv3 features and Flux2 VAE latent	CONDITIONING
negative	Negative conditioning data containing zero-filled DINOv3 features and zero-filled Flux2 VAE latent	CONDITIONING
latent	The fixed size noise target (latent sequence plus camera token) for the KSampler	LATENT

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): 9187a4a020818b9adc762eb41e913086b59d62c47abe92d4bafdb14bc8779f51