ArtVenture Custom Nodes
A comprehensive set of custom nodes for ComfyUI, focusing on utilities for image processing, JSON manipulation, model operations and working with object via URLs
Image Nodes
LoadImageFromUrl
Loads images from URLs.
Inputs:
image
: List of URLs or base64 image data, separated by new lineskeep_alpha_channel
: Preserve alpha channeloutput_mode
: List or batch output. UseList
if you have different resolutions.
JSON Nodes
LoadJsonFromUrl
Loads JSON data from URLs.
Inputs:
url
: JSON URLprint_to_console
: Print JSON to console
LoadJsonFromText
Loads JSON data from text.
Inputs:
data
: JSON textprint_to_console
: Print JSON to console
Get<*>FromJson
Includes GetObjectFromJson
, GetTextFromJson
, GetFloatFromJson
, GetIntFromJson
, GetBoolFromJson
.
Use key format key.[index].subkey.[sub_index]
to access nested objects.
Utility Nodes
StringToNumber
Converts strings to numbers.
Inputs:
string
: Input stringrounding
: Rounding method
TextRandomMultiline
Randomizes the order of lines in a multiline string.
Inputs:
text
: Input textamount
: Number of lines to randomizeseed
: Random seed
Inpainting Nodes
PrepareImageAndMaskForInpaint
Prepares images and masks for inpainting operations. It's to mimic the behavior of the inpainting in A1111.
Inputs:
image
: Input image tensormask
: Input mask tensormask_blur
: Blur amount for mask (0-64)inpaint_masked
: Whether to inpaint only the masked regions, otherwise it will inpaint the whole image.mask_padding
: Padding around mask (0-256)width
: Manually set inpaint area width. Leave 0 default to the masked area plus padding. (0-2048)height
: Manually set inpaint area height. (0-2048)
Outputs:
inpaint_image
: Processed image for inpaintinginpaint_mask
: Processed maskoverlay_image
: Preview overlaycrop_region
: Crop coordinates (input of OverlayInpaintedImage)
OverlayInpaintedImage
Overlays inpainted images with original images.
Inputs:
inpainted
: Inpainted imageoverlay_image
: Original imagecrop_region
: Crop region coordinates
Outputs:
IMAGE
: Final composited image
LaMaInpaint
Remove objects from images using LaMa model.
LLM Nodes
LLMApiConfig
Configures generic LLM API parameters.
Inputs:
model
: Model name (GPT-3.5, GPT-4, etc)max_token
: Maximum tokenstemperature
: Temperature parameter
OpenAIApi
Configures OpenAI API access.
Inputs:
openai_api_key
: OpenAI API keyendpoint
: API endpoint URL
Claude API Nodes
ClaudeApi
Configures Anthropic Claude API access.
Inputs:
claude_api_key
: Claude API keyendpoint
: API endpointversion
: API version
AwsBedrockClaudeApi
Configures AWS Bedrock Claude API access.
Inputs:
aws_access_key_id
: AWS access keyaws_secret_access_key
: AWS secret keyregion
: AWS regionversion
: API version
AwsBedrockMistralApi
Configures AWS Bedrock Mistral API access.
Inputs:
aws_access_key_id
: AWS access keyaws_secret_access_key
: AWS secret keyregion
: AWS region
LLMMessage
Creates a message for LLM interaction.
Inputs:
role
: Message role (system/user/assistant)text
: Message contentimage
: Optional image inputmessages
: Previous message history
LLMChat
Handles chat interactions with LLMs.
Inputs:
messages
: Message historyapi
: LLM API configurationconfig
: Model configurationseed
: Random seed
LLMCompletion
Handles completion requests to LLMs.
Inputs:
prompt
: Input promptapi
: LLM API configurationconfig
: Model configurationseed
: Random seed