Node Overview
Nodes are essential components in a workflow, enabling a sequence of operations by connecting nodes with various functionalities.
Core Nodes
Start: Sets the initial parameters to begin a workflow process.
End: Defines the final output to conclude a workflow process.
Answer: Specifies response content in a Chatflow process.
Large Language Model (LLM): Utilizes a large language model for answering questions or processing natural language.
Knowledge Retrieval: Fetches relevant text content from a knowledge base for context in downstream LLM nodes.
Question Classifier: Uses defined classification descriptions to allow the LLM to categorise user input.
IF/ELSE: Splits the workflow into two branches based on conditional statements.
Code Execution: Executes custom logic, such as data transformation, using Python/NodeJS code.
Template Transformation: Uses Jinja2, a Python templating language, for flexible data transformation and text processing.
Variable Aggregator: Combines variables from multiple branches into one for unified configuration in downstream nodes.
Parameter Extractor: Uses LLM to infer and extract structured parameters from natural language for subsequent tool usage or HTTP requests.
Iteration: Repeats steps on list objects until all results are processed.
HTTP Request: Sends server requests using the HTTP protocol for external data retrieval, webhooks, image generation, and more.
Tools: Allows the integration of built-in Dify tools, custom tools, sub-workflows, and more within the workflow.
Last updated