ComfyUI Node: Save Image (16 bit)

Save Image (16 bit):

The SaveImageHighPrec node is designed to save images in high precision 16-bit PNG format, ensuring that the fine details and color depth of your images are preserved. This node is particularly useful for AI artists who require high-quality image outputs for further processing or final presentation. By leveraging this node, you can maintain the integrity of your images' color information, which is crucial for tasks that demand high fidelity and precision. The node handles the conversion and saving process seamlessly, allowing you to focus on your creative work without worrying about technical details.

Save Image (16 bit) Input Parameters:

images

This parameter expects a list of images that you want to save. Each image should be in a tensor format, typically generated or processed by other nodes in your workflow. The images are saved in 16-bit PNG format, which provides higher color depth and precision compared to standard 8-bit images. This is particularly important for preserving subtle gradients and details in your artwork.

filename_prefix

This parameter allows you to specify a prefix for the filenames of the saved images. The prefix helps in organizing and identifying your saved files, especially when dealing with multiple images. The default value is "ComfyUI", but you can customize it to suit your project needs. The node appends a unique identifier to each filename to prevent overwriting existing files.

prompt

This hidden parameter is used to include additional metadata in the saved PNG files. It can store information related to the image generation process, such as the prompt used for generating the image. This metadata can be useful for tracking and reproducing your work.

extra_pnginfo

This hidden parameter allows you to add extra PNG information chunks to the saved files. These chunks can contain various types of metadata, such as creation date, software used, or any other custom information you want to embed in the PNG files. This can be useful for archival purposes or for providing additional context to the images.

Save Image (16 bit) Output Parameters:

ui

This output parameter provides a dictionary containing information about the saved images. The dictionary includes details such as the filenames, subfolders, and the type of images saved. This information can be used to reference the saved files in subsequent nodes or for further processing. It helps in keeping track of the saved images and their locations, ensuring a smooth workflow.

Save Image (16 bit) Usage Tips:

• Ensure that your images are in the correct tensor format before passing them to the SaveImageHighPrec node to avoid errors during the saving process.
• Use a meaningful filename prefix to easily identify and organize your saved images, especially when working on multiple projects.
• Take advantage of the prompt and extra_pnginfo parameters to embed useful metadata in your PNG files, which can help in tracking and reproducing your work.

Save Image (16 bit) Common Errors and Solutions:

ColorMod: Only PNG files can be loaded in 16 bit color.

• Explanation: This error occurs when attempting to load a non-PNG file in 16-bit color mode.
• Solution: Ensure that the file you are trying to load is in PNG format. Convert other file formats to PNG before loading them.

Failed to read image!

• Explanation: This error indicates that the node was unable to read the specified image file.
• Solution: Check the file path and ensure that the image file exists and is accessible. Verify that the file is not corrupted and is in the correct format.

Invalid image file: <filename>

• Explanation: This error occurs when the specified image file does not exist or is not valid.
• Solution: Verify the file path and ensure that the image file is present and correctly named. Check for any typos or incorrect directory paths.

AssertionError: Image tensor must have 4 dimensions (batch, height, width, channels)

• Explanation: This error occurs when the input image tensor does not have the required dimensions.
• Solution: Ensure that your image tensor has the correct dimensions before passing it to the node. The tensor should have four dimensions: batch size, height, width, and channels.