FlowiseAI Integration

Overview

FlowiseAI is a powerful low-code platform for building custom AI workflows and agents with an intuitive visual interface. It enables you to create sophisticated AI applications without extensive programming knowledge.

Key Features

Visual Node-Based Editor: Drag-and-drop interface for creating AI workflows
Pre-built Components: Extensive library of ready-to-use AI components
Multi-Modal Support: Text, image, and audio processing capabilities
Custom Function Support: Extend functionality with JavaScript functions

Use Cases

Building conversational AI agents for customer service
Creating data processing and analysis pipelines
Developing content generation and summarization tools
Integrating multiple AI services into cohesive workflows

Setup Instructions

1. Install FlowiseAI

Visit the FlowiseAI GitHub repo for installation options, either via Docker or NPM:

docker pull flowiseai/flowise
docker run -d --name flowise -p 3000:3000 flowiseai/flowise

npm install -g flowise
npx flowise start

2. Access FlowiseAI Dashboard

Open your browser and go to http://localhost:3000. You’ll see the FlowiseAI dashboard with the canvas area.

FlowiseAI Dashboard Home

3. Create a New Flow

Click on the + Add New button to create a new flow. Give your flow a name like relaxAI Integration.

4. Navigate to LangChain Components

In the components panel on the left, navigate to the LangChain section. Find the ChatModels category.

5. Add ChatModel Node

Drag the ChatOpenAI Custom ChatModel node to your canvas. This will be your connection point to relaxAI.

6. Select OpenAI Compatible Option

In the node configuration panel on the right, select Custom OpenAI API Compatible from the Model Name dropdown.

Adding ChatModel Node

7. Configure relaxAI Settings

Enter the following details in the configuration panel:

Base URL: https://api.relax.ai/v1/
API Key: RELAX_API_KEY
Model Name: <model name> # for example: Llama-4-Maverick-17B-128E

Configuration Parameters

8. Set Additional Parameters

Configure optional parameters like Temperature, Top P, and Max Tokens
Recommended starting values:
- Temperature: 0.7
- Top P: 0.95

9. Complete Your Flow

Add additional components like ChatPrompt, TextInput, and TextOutput

Connect the components to create a complete flow:

TextInput → ChatPrompt
ChatPrompt → ChatModel (relaxAI)
ChatModel → TextOutput

Complete Flow Example

10. Save Your Flow

Click the “Save” button in the top-right corner
Your relaxAI integration is now ready to use

11. Test Your Setup

Click the “Prediction” tab at the top
Enter a test message and check if your relaxAI model responds correctly

Code Example

import { ChatOpenAI } from "@langchain/openai";

const llm = new ChatOpenAI({
  modelName: "Llama-4-Maverick-17B-128E",
  openAIApiKey: "RELAX_API_KEY",
  configuration: {
    baseURL: "https://api.relax.ai/v1/"
  }
});