StableZero123_Conditioning - ComfyUI Built-in Node Documentation

The StableZero123Conditioning node processes an input image and camera angles to generate conditioning data and latent representations for 3D model generation.

The StableZero123_Conditioning node processes an input image and camera angles to generate conditioning data and latent representations for 3D model generation. It uses a CLIP vision model to encode the image features, combines them with camera embedding information based on elevation and azimuth angles, and produces positive and negative conditioning along with a latent representation for downstream 3D generation tasks.

Inputs

Parameter	Description	Data Type	Required	Range
clip_vision	The CLIP vision model used to encode image features	CLIP_VISION	Yes	-
init_image	The input image to be processed and encoded	IMAGE	Yes	-
vae	The VAE model used for encoding pixels to latent space	VAE	Yes	-
width	Output width for the latent representation (default: 256, step: 8)	INT	Yes	16 to MAX_RESOLUTION
height	Output height for the latent representation (default: 256, step: 8)	INT	Yes	16 to MAX_RESOLUTION
batch_size	Number of samples to generate in the batch (default: 1)	INT	Yes	1 to 4096
elevation	Camera elevation angle in degrees (default: 0.0, step: 0.1)	FLOAT	Yes	-180.0 to 180.0
azimuth	Camera azimuth angle in degrees (default: 0.0, step: 0.1)	FLOAT	Yes	-180.0 to 180.0

Note: The width and height parameters use a step of 8, so values are set in increments of 8. The node divides them by 8 to determine the latent representation dimensions. The image is rescaled to the given width and height using bilinear upscaling with center cropping before VAE encoding.

Outputs

Output Name	Description	Data Type
positive	Positive conditioning data combining image features and camera embeddings, including the VAE-encoded input image as a latent to concatenate	CONDITIONING
negative	Negative conditioning data with zero-initialized features and a zero-initialized latent	CONDITIONING
latent	Zero-initialized latent representation with dimensions [batch_size, 4, height//8, width//8]	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): a694610c9f22fe0dab3ae02f4aabb33e3de8e5031c82dff5e8ba232c098f4a1d