YellowBus (klinter)

Documentation

• Class name: YellowBus
• Category: None
• Output node: False

YellowBus is designed to facilitate the transfer of various types of data, including models, VAEs, conditioning prompts, and latent embeddings, without modification. It acts as a conduit, allowing these elements to flow seamlessly from input to output, thereby enabling the preservation and straightforward exchange of information between different stages of a processing pipeline.

Input types

Required

• model
  • Represents a model input, allowing for the transfer of model data through the node.
  • Comfy dtype: MODEL
  • Python dtype: object
• vae
  • Represents a VAE (Variational Autoencoder) input, facilitating the transfer of VAE data through the node.
  • Comfy dtype: VAE
  • Python dtype: object
• pos_prompt
  • Represents a positive conditioning prompt, enabling the transfer of specific conditioning information intended to positively influence the outcome.
  • Comfy dtype: CONDITIONING
  • Python dtype: str
• neg_prompt
  • Represents a negative conditioning prompt, enabling the transfer of specific conditioning information intended to negatively influence the outcome.
  • Comfy dtype: CONDITIONING
  • Python dtype: str
• latent
  • Represents latent embeddings input, allowing for the transfer of latent space representations through the node.
  • Comfy dtype: LATENT
  • Python dtype: numpy.ndarray

Output types

• model
  • Comfy dtype: MODEL
  • Outputs the model data received as input, unchanged.
  • Python dtype: object
• vae
  • Comfy dtype: VAE
  • Outputs the VAE (Variational Autoencoder) data received as input, unchanged.
  • Python dtype: object
• pos_prompt
  • Comfy dtype: CONDITIONING
  • Outputs the positive conditioning prompt received as input, unchanged.
  • Python dtype: str
• neg_prompt
  • Comfy dtype: CONDITIONING
  • Outputs the negative conditioning prompt received as input, unchanged.
  • Python dtype: str
• latent
  • Comfy dtype: LATENT
  • Outputs the latent embeddings received as input, unchanged.
  • Python dtype: numpy.ndarray

Usage tips

• Infra type: CPU
• Common nodes: unknown

Source code

class YellowBus:
    @classmethod
    def INPUT_TYPES(cls):
        return {
            "required": {
                "model": ("MODEL", {"forceInput": False,}),         # Input is a model
                "vae": ("VAE", {"forceInput": False,}),             # Input is a VAE
                "pos_prompt": ("CONDITIONING", {"forceInput": False,}),  # Input is a positive prompt (conditioning)
                "neg_prompt": ("CONDITIONING",  {"forceInput": False,}),  # Input is a negative prompt (conditioning)
                "latent": ("LATENT", {"forceInput": False,}),       # Input is latent embeddings
            }
        }

    RETURN_TYPES = ("MODEL", "VAE", "CONDITIONING", "CONDITIONING", "LATENT")
    RETURN_NAMES = ("model", "vae", "pos_prompt", "neg_prompt", "latent")
    FUNCTION = "transfer"

    def transfer(self, model, vae, pos_prompt, neg_prompt, latent):
        # Return inputs directly as outputs
        return (model, vae, pos_prompt, neg_prompt, latent)