Skip to main content
FlowiseAI is a low-code platform for building LLM applications visually. Use the native ChatCometAPI node to connect CometAPI and build chatbots, Q&A systems, and agent workflows.

​
Prerequisites

1

Get your CometAPI API key

Log in to the CometAPI console. Click Add API Key and copy your sk-xxxxx key.
CometAPI dashboard showing the Add API Key button
CometAPI API key details with base URL
2

Add the ChatCometAPI node

  1. In your FlowiseAI chatflow canvas, click Add New.
FlowiseAI canvas with the Add New button
  1. Search for cometapi and select the ChatCometAPI node.
Node search showing ChatCometAPI result
  1. Drag it onto the canvas.
ChatCometAPI node placed on the canvas
  1. In the node’s Connect Credential dropdown, select Create New, enter your sk-xxxxx key, and save.
Credential dialog with API key entered
Optionally set Model Name to the current model ID you want to use and Temperature (e.g. 0.7).
3

Build the LLM Chain

  1. Add a Prompt Template node (search “Prompt”) and configure a template, e.g.: 
    Question: {question}
Answer: Let's think step by step.
Prompt Template node with the example template
Prompt Template node placed on canvas
  1. Add an LLM Chain node (search “Chain”). Connect:
    • Language Model input ← ChatCometAPI node
    • Prompt input ← Prompt Template node
  2. Give the chain a name, e.g. Universal Chat Assistant.
LLM Chain node with Language Model and Prompt connected
Complete chatflow with all three nodes connected
4

Connect nodes and save

Ensure the chain forms a complete flow: Input → Prompt Template → LLM Chain → Output. Click Save Chatflow.
Complete chatflow canvas with all connections visible
  • Cannot find the “cometapi” node — ensure your FlowiseAI version supports this integration. Check for updates or relevant plugins.
  • Credential configuration failed — verify the API key is correct and has not expired. Copy it fresh from the CometAPI console.
  • Node connection error — confirm the connection lines point to the correct inputs (Language Model and Prompt). If the response quality is poor, adjust the Prompt Template or lower the Temperature value.