Skip to main content
Flowise provides a rich ecosystem of integrations to build powerful AI applications. Each integration is designed to work seamlessly within the visual workflow builder.

​
Categories

Language Models

Connect to OpenAI, Anthropic, Google, Azure, and local models like Ollama

Vector Stores

Store and retrieve embeddings using Pinecone, Chroma, Qdrant, and more

Document Loaders

Load data from PDFs, websites, text files, and 30+ other sources

Tools & Agents

Extend AI capabilities with calculators, web browsers, and custom tools

​
Quick Start

1

Choose Your Integration

Browse available integrations in the Flowise canvas sidebar, organized by category
2

Add Credentials

Configure API keys and credentials in the Credentials section
3

Drag and Connect

Add nodes to your canvas and connect them to build your workflow
4

Configure Parameters

Adjust settings like temperature, chunk size, and model selection

​
Language Models

  • OpenAI - GPT-4, GPT-3.5, with function calling and vision support
  • Anthropic - Claude 3.5 Sonnet with extended thinking capabilities
  • Google - Gemini 1.5 Pro/Flash with multimodal support
  • Ollama - Run Llama 2, Mistral, and other models locally

​
Vector Stores

  • Pinecone - Fully managed vector database with high performance
  • Chroma - Open-source embedding database for local development
  • Qdrant - Scalable vector search written in Rust
  • Weaviate - Open-source vector database with hybrid search

​
Document Loaders

  • PDF - Extract text from PDF files with page splitting
  • Web Scraper - Load content from websites with CSS selectors
  • API - Load data from REST APIs
  • File Loaders - Support for 30+ file formats

​
Integration Features

All integrations support:
  • Visual configuration in the canvas
  • Environment variable support
  • Metadata filtering and customization
  • Batch processing where applicable

​
Common Parameters

Most integrations share these common configuration options:
​
credential
credential
required
API key or authentication credential for the service
​
cache
BaseCache
Optional caching layer to reduce API calls and improve performance
​
metadata
json
Additional metadata to attach to documents or requests

​
Best Practices

  1. Use Environment Variables - Store API keys in environment variables rather than hardcoding
  2. Enable Caching - Reduce costs by caching LLM responses
  3. Set Appropriate Limits - Configure timeouts and rate limits
  4. Monitor Usage - Track API usage to avoid unexpected costs
  5. Test Locally - Use Ollama or local models during development

​
Need Help?

Each integration page includes:
  • Configuration parameters and descriptions
  • Code examples and usage patterns
  • Common use cases and troubleshooting tips
  • Links to official documentation
Explore the integration categories to learn more about each available integration.

Build docs developers (and LLMs) love

Get started for freeTalk to us