# Flowise

> Low-code tool for building LLM apps with drag-and-drop.

- Category: AI & Machine Learning
- Difficulty: beginner
- Components: flowise
- Tags: flowise, llm, low-code, langchain
- Source: https://docker.recipes/ai-ml/flowise

## Overview

Flowise is an open-source low-code platform that enables users to build Language Model applications through an intuitive drag-and-drop interface. Created as a visual workflow builder for LangChain and LlamaIndex, Flowise democratizes AI application development by allowing non-technical users to create sophisticated chatbots, document Q&A systems, and AI agents without writing code. The platform supports integration with major AI providers including OpenAI, Azure OpenAI, HuggingFace, Anthropic, and local models through Ollama.

This Docker deployment creates a self-contained Flowise instance with persistent storage for your chatflows, credentials, and configuration. The container exposes a web interface where you can design AI workflows by connecting nodes representing different components like language models, vector stores, memory systems, and tools. Built-in authentication protects your instance while volume mounting ensures your created workflows survive container restarts.

This setup is ideal for AI enthusiasts, rapid prototyping teams, and organizations wanting to experiment with Large Language Models without complex infrastructure. Flowise bridges the gap between powerful AI capabilities and accessible implementation, making it perfect for proof-of-concepts, internal tools, and customer-facing AI applications that can be deployed as REST APIs.

## Features

- Visual drag-and-drop interface for building LangChain and LlamaIndex workflows without coding
- Built-in marketplace with pre-configured templates for common AI use cases like customer support and document analysis
- Multi-model support including OpenAI GPT, Claude, local Ollama models, and HuggingFace transformers
- Integrated vector database connections for Pinecone, Weaviate, Chroma, and FAISS document retrieval
- Export chatflows as REST API endpoints for direct integration with external applications
- Memory management systems including buffer memory, conversation summary, and entity memory
- Real-time chat testing interface with conversation history and debug information
- Credential management system for securely storing API keys and connection strings

## Use cases

- Building customer support chatbots with company document knowledge bases using RAG patterns
- Creating internal AI assistants for employee onboarding and FAQ automation
- Prototyping conversational AI applications before committing to custom development
- Educational environments for teaching AI concepts and LangChain workflows visually
- Small business automation for lead qualification and appointment scheduling chatbots
- Content creation workflows combining multiple AI models for research and writing assistance
- Document analysis systems for processing contracts, invoices, and compliance materials

## Prerequisites

- Docker and Docker Compose installed with at least 2GB available RAM for model inference
- API keys for chosen AI providers (OpenAI, Anthropic, etc.) or local Ollama installation
- Port 3000 available on the host system for web interface access
- Basic understanding of AI concepts like embeddings, vector databases, and prompt engineering
- Administrative credentials defined in environment variables for secure access
- Sufficient disk space for vector embeddings and conversation history storage

## docker-compose.yml

```yaml
services:
  flowise:
    image: flowiseai/flowise:latest
    container_name: flowise
    restart: unless-stopped
    environment:
      PORT: 3000
      FLOWISE_USERNAME: ${ADMIN_USER}
      FLOWISE_PASSWORD: ${ADMIN_PASSWORD}
    volumes:
      - flowise_data:/root/.flowise
    ports:
      - "3000:3000"

volumes:
  flowise_data:
```

## .env

```bash
ADMIN_USER=admin
ADMIN_PASSWORD=changeme
```

## Usage notes

- Docs: https://docs.flowiseai.com/
- Access at http://localhost:3000 - login with env credentials
- Drag-and-drop LangChain/LlamaIndex workflow builder
- Connect to OpenAI, Ollama, HuggingFace, Azure, and more
- Export chatflows as API endpoints for integration
- Marketplace has pre-built templates for common use cases

## Troubleshooting

- Authentication failed or blank login page: Verify FLOWISE_USERNAME and FLOWISE_PASSWORD environment variables are properly set
- Chatflow fails with API rate limit errors: Check your AI provider API key quotas and implement retry mechanisms in your flows
- Vector similarity search returns poor results: Ensure document chunking size matches your embedding model's context window
- Container startup fails with permission errors: Check that the flowise_data volume has proper read/write permissions for the container user
- Memory errors during large document processing: Increase Docker container memory limits and reduce document batch sizes
- Exported API endpoints return 404 errors: Verify the chatflow is properly saved and deployed through the Flowise interface before API access

