Initializes and returns a logger instance for an individual custom node module. It sets the logger's level to match the parent package logge

Validates if a given string is a valid GCS URI and if the object it points to exists and is identified as an image. Args: gcs_ur

Processes the video generation operation response and saves generated videos. Args: operation: The completed LRO (Long Running Opera

Converts a PyTorch tensor or PIL Image into PNG-encoded bytes. This function processes an input image, which can be either a PyTorch tensor o

Generates content using the Gemini API based on the provided prompt and parameters. This method constructs a request to the Gemini API, inclu

* Manages a single video preview widget for a given component, with navigation for multiple videos. * @param {object} targetComponent The component t

Initializes the Gemini client. Args: gcp_project_id: The GCP project ID. If provided, overrides metadata lookup.

Generates an image using the Gemini Flash Image model. Args: model: The name of the Gemini model to use. aspect_ratio

Upscale an image using Imagen4. Args: model: Imagen4 model id. There are three as of Jul 1, 2025. image: The

Initializes the Gemini client. Args: gcp_project_id: The GCP project ID. If provided, overrides metadata lookup.

Initializes the Gemini client. Args: gcp_project_id: The GCP project ID. If provided, overrides metadata lookup.

Makes a prediction using the Vertex AI PredictionServiceClient. This method is decorated with `@api_error_retry` to handle API error

Downloads a file from a Google Cloud Storage (GCS) URI to a specified local destination path. This function validates the GCS URI format, ex

Generates content using the Gemini 3.5 API based on the provided prompt and parameters. Args: prompt (str): The main text prompt

Generates an image using the Gemini Pro Image model. Args: model: The name of the Gemini model to use. default: gemini-3-pro-imag

Generate image from text prompt using Imagen4. Args: model: Imagen4 model id. There are three as of Jul 1, 2025.

Generate image from text prompt using Imagen3. Args: prompt: The text prompt for image generation. person_ge

Generates music from a text prompt using the Lyria 2 API. Args: prompt: The text prompt for music generation.

Generates video from a Google Cloud Storage (GCS) image URI using the Veo 3.1 API. Args: model: Veo3 model.

Generates video from a Google Cloud Storage (GCS) image URI using the Veo 2.0 API. Args: gcsuri: The GCS URI of the inpu

Generates video from an image input (as a torch.Tensor) using the Veo 3.1 API. Args: model: Veo3 model. imag

Generates video from an image input (as a torch.Tensor) using the Veo 2.0 API. Args: image: The input image as a torch.T

Generates a video from the references. Args: model: Veo3 model. prompt: The text prompt for video generation

Generates video from a text prompt using the Veo 3.1 API. Args: model: Veo3 model. prompt: The text prompt f

Generates video from a text prompt using the Veo 2.0 API. Args: prompt: The text prompt for video generation.

Reads a media file (image, audio, or video) and converts it to a genai.types.Part. This function is designed to prepare the raw bytes of a media

Reads the custom format, level, handler class, and arguments from logging.conf, creates a custom isolated handler, and attaches it to the pac

Converts a PyTorch tensor or PIL Image into PNG-encoded bytes. This function processes an input image, which can be either a PyTorch tensor o

Defines the input types and widgets for the ComfyUI node. Returns: A dictionary specifying the required and optional inp

Defines the input types and widgets for the ComfyUI node. Returns: A dictionary specifying the required and optional inp

Defines the input types for the node. - video_paths: A list where each item is a path to a video file.

Defines the input types and widgets for the ComfyUI node. Returns: A dictionary specifying the required and optional inp

Defines the input types for the Veo3ReferenceToVideo node.

Defines the input types and widgets for the ComfyUI node. Returns: A dictionary specifying the required and optional inp

Defines the input types and widgets for the ComfyUI node. Returns: A dictionary specifying the required and optional inp

Defines the input types and widgets for the ComfyUI node. Returns: A dictionary specifying the required and optional inp

Defines the input types and widgets for the ComfyUI node. Returns: A dictionary specifying the required and optional inp

Initializes the ImagenTextToImageNode.

Initializes the ImagenUpscaleImageNode.

Initializes the Imagen4API client. Args: project_id: The GCP project ID. If None, it will be retrieved from GCP metadata

Initializes the Vertex AI client. Args: gcp_project_id: The GCP project ID. If provided, overrides metadata lookup.

Initializes the Imagen3API client. Args: project_id: The GCP project ID. If None, it will be retrieved from GCP metadata

Initializes the Gemini 3 Pro Image Preview client. Args: project_id (Optional[str], optional): The GCP project ID. If not provided

Initializes the Veo3API client. Args: project_id: The GCP project ID. If None, it will be retrieved from GCP metadata.

Initializes the client. Args: project_id: The GCP project ID. If provided, overrides metadata lookup. region

Initializes the Gemini3ProImage Node.

Initializes the Gemini 2.5 Flash Image Preview client. Args: project_id (Optional[str], optional): The GCP project ID. If not prov

Initializes the Gemini 3.5 client. Args: project_id: The GCP project ID. If provided, overrides metadata lookup. regio

Initializes the ImagenTextToImageNode.

Initializes the Gemini31FlashImage Node.

Initializes the Veo2API client. Args: project_id: The GCP project ID. If None, it will be retrieved from GCP metadata.

Initializes the Imagen3TextToImageNode.

Decorator factory that implements centralized API error handling, retry logic, and exception mapping for functions interacting with the Googl

Decodes a base64 string and converts the resulting image to a PyTorch tensor. Args: base64_string: The base64 encoded string of an i

Loads multiple videos from newline-separated paths, extracts frames, and returns them as a single IMAGE batch.

Generates a video from a text prompt using the Google Veo 2.0 API. Args: prompt: The text prompt for video generation.

Generates a video from a GCS image URI using the Google Veo 2.0 API. Args: gcsuri: The GCS URI of the input image.

Generates a video from an input image (torch.Tensor) using the Google Veo 2.0 API. Args: image: The input image as a tor

Generates a video from a text prompt using the Google Veo 3.0 API. Args: model: Veo3 model prompt: The text

Generates a video from a GCS image URI using the Google Veo 3.0 API. Args: model: Veo3 model gcsuri: The GCS

Generates a video from an input image (torch.Tensor) using the Google Veo 3.0 API. Args: model: Veo3 model i

Generates images based on the provided parameters using the Imagen API and returns them as a PyTorch tensor suitable for ComfyUI.

This function iterates through each provided product image, makes an API call to the Virtual Try-On service, and returns the generate

Generates images using the Gemini Pro Image API and returns them. This method interfaces with the GeminiProImageAPI to generate images

Generates images using the Gemini Flash Image API and returns them. This method interfaces with the GeminiFlashImageAPI to generate images

Generates images using the Gemini Flash Image API and returns them. This method interfaces with the GeminiFlashImageAPI to generate images

Generates images based on the provided parameters using the Imagen API and returns them as a PyTorch tensor suitable for ComfyUI.

Generates content using the Gemini API based on the provided prompt and parameters. This method constructs a request to the Gemini API, inclu

Generate image from text prompt using Imagen. Args: client: genai.Client model: model to be used prompt: The text pr

Generates music from a text prompt using the Lyria 2 API. Args: prompt: The text prompt for music generation.

Generates video from a Google Cloud Storage (GCS) image URI using the Veo API. Args: client: genai.Client model: model to be

Generates video from an image input (as a torch.Tensor) using the Veo API. Args: client: genai.Client model: model to be use

Generates video from GCS reference image URIs using the Veo API.

Generates video from a text prompt using the Veo API. Args: client: genai.Client model: model to be used prompt: The

Attempts to prepare a media file into a genai.types.Part for input to the model. This function checks if the specified file exists and, if so, at