🔀 Setup Flowise — No-Code AI Pipeline Builder
Deploy Flowise on Ubuntu with Docker — a drag-and-drop LangChain pipeline builder for creating AI chatbots, agents, and RAG workflows visually. Connect Ollama, OpenAI, or any LLM and build production-ready AI applications without writing code.
📦 Resources & Setup Scripts
Grab the automated bash script from GitHub to follow along with the video.
Quick Install:
wget https://raw.githubusercontent.com/mhmdali94/Docker/main/ai/flowise/flowise-ubuntu.sh
chmod +x flowise-ubuntu.sh
sudo bash flowise-ubuntu.shTutorial Steps
1 Download & Run the Script
The script installs Docker and deploys Flowise with a persistent storage volume for your saved chatflows and credentials.
wget https://raw.githubusercontent.com/mhmdali94/Docker/main/ai/flowise/flowise-ubuntu.sh
chmod +x flowise-ubuntu.sh
sudo bash flowise-ubuntu.sh2 Access the Flowise UI
Open your browser and navigate to the Flowise interface:
http://<your-server-ip>:30003 Add Your LLM Credentials
Go to Credentials in the sidebar and add your API keys — OpenAI, Anthropic, or your Ollama server URL. Credentials are encrypted and stored locally.
4 Build Your First Chatflow
Click "Add New" to create a chatflow. Drag nodes from the left panel — choose an LLM node, a memory node, and a chat input — connect them, and click Save. Your chatflow is instantly available as an API endpoint and embeddable chat widget.
Ports Used
| Port | Purpose |
|---|---|
| 3000 | Flowise Web UI & API |
Overview
Flowise is an open-source, low-code platform for building LLM-powered applications using a visual drag-and-drop interface built on LangChain and LlamaIndex. It abstracts the complexity of chaining LLMs, vector stores, memory modules, and tool integrations into a node-based canvas. Each saved 'chatflow' is automatically published as a REST API endpoint, a shareable chat UI, and an embeddable widget — making it one of the fastest ways to go from idea to deployed AI application without writing server code.
Why Use It
Flowise lowers the barrier to building LangChain applications from requiring Python expertise to drag-and-drop configuration. For developers who understand AI concepts but do not want to maintain complex Python codebases, Flowise offers the same LangChain capabilities through a visual interface. Its killer feature is the immediate API exposure — every chatflow is an endpoint — which means you can build an AI backend for your application in minutes. Its lightweight single-container architecture also makes it the easiest self-hosted AI builder to deploy and maintain.
When You Need It
Who Should Use It
Real Use Cases
Main Features
How to Use After Installation
Security Best Practices
Ports and Firewall Notes
Flowise runs on port 3000 for both the UI and API. Block this port at the firewall and expose it only through a reverse proxy with HTTPS. Individual chatflow API endpoints do not require authentication by default — anyone with the endpoint URL can query them. Enable API key protection per chatflow under the chatflow's configuration panel before sharing endpoints externally.
Backup and Maintenance
Common Mistakes
Troubleshooting
Alternatives
Dify offers a more enterprise-grade experience with team workspaces, fine-grained access controls, and a more polished app publishing workflow — better for teams building multiple production applications. n8n is a better choice when AI is one component of a broader automation involving dozens of external services and APIs. LangFlow is Flowise's closest competitor with a similar node-based canvas — slightly more aligned with pure LangChain workflows. AnythingLLM is simpler if your only need is document Q&A without building custom pipelines. For maximum control, LangChain or LlamaIndex Python code gives the same functionality without the UI constraint.
When Not to Use It
Flowise is not ideal for high-throughput production AI APIs that need to handle hundreds of concurrent requests — it is optimized for team-scale use. If you need complex conditional branching, dynamic routing between many LLM providers, or deep observability with traces and logs, Dify or LangSmith are better fits. If your use case involves batch processing large document sets automatically, a Python script with LangChain is more maintainable than a visual chatflow. Flowise also lacks native multi-tenant isolation — running it for multiple separate clients on one instance requires careful API key management.
Need Help Setting Up Flowise?
PrismaTechWork provides end-to-end infrastructure services — from initial deployment and security hardening to ongoing monitoring, automated backups, and dedicated support. Whether you need a single-server setup or a multi-site network, our team ensures your infrastructure is built right, secured properly, and maintained reliably.
Frequently Asked Questions
What is the difference between Flowise and Dify?
Flowise is a lightweight, single-container LangChain visual builder focused on chatflows and agent pipelines. Dify is a heavier, multi-container platform with a stronger focus on app publishing, team collaboration, and enterprise features like SSO and annotation tools. Flowise is easier to deploy and faster to get started with. Dify provides more production-grade features for teams shipping multiple AI applications. Choose Flowise for prototyping speed; choose Dify when you need a production app platform.
Can I use Flowise with a local Ollama model?
Yes. Add your Ollama server URL under Credentials (e.g., http://host.docker.internal:11434 if on the same Docker host, or http://your-server-ip:11434 for a remote server). In a chatflow, drag an 'ChatOllama' node from the node panel, select your credential, and choose the model name. Flowise will use Ollama for inference without any cloud API calls.
How do I expose a chatflow as an API?
Every saved chatflow automatically gets a REST API endpoint. Click the '</>' (API Endpoint) button in the chatflow editor toolbar to see the endpoint URL and example curl command. By default the endpoint is public — anyone with the URL can query it. To require an API key, go to the chatflow's configuration and enable 'Chatflow Config → API Key'. Flowise also generates a React component and a script embed tag for website integration.
Does Flowise support conversation memory?
Yes. Flowise has multiple memory node types — Buffer Memory (keeps the full conversation history), Buffer Window Memory (keeps the last N messages), Summary Memory (uses an LLM to compress older history), and vector store-backed memory for long-term recall. Drag a memory node into your chatflow and connect it to the LLM node. Memory is session-scoped by default, meaning each API session ID gets its own conversation history.
Can Flowise connect to external databases or APIs?
Yes, through tool nodes and custom function nodes. Flowise has built-in tool nodes for HTTP requests, web search (SerpAPI, Google), a calculator, and weather APIs. You can also add a 'Custom Tool' node with JavaScript code to call any external API or database. For database connections, use the custom function node with a database client library included in the Flowise container, or add a custom Docker layer.
Is Flowise suitable for multi-user teams?
Flowise supports basic multi-user access through username/password authentication and API key management. However, it does not have workspace isolation — all users with login access can see and edit all chatflows. For teams needing per-user or per-department isolation of chatflows and credentials, Dify or a separate Flowise instance per team is a better model. Flowise works well for small technical teams that trust each other with shared access.
How do I add documents to a RAG chatflow in Flowise?
In your chatflow, add a Document Loader node (PDF, DOCX, web URL, or folder), connect it to a Text Splitter node, then connect to a Vector Store node (Chroma, Pinecone, Qdrant, etc.), and finally connect the vector store to a Retrieval QA Chain node backed by your LLM. Click the Upsert button on the vector store node to index the documents. After upsert completes, the chatflow can retrieve and answer questions from your documents.
How do I update Flowise without losing my chatflows?
Run `docker compose pull && docker compose up -d` in your Flowise directory to pull the latest image. Your chatflows, credentials, and vector stores are stored in the named Docker volume and are not affected by image updates. Before major version upgrades, export important chatflows as JSON from the UI as an extra backup. Check the Flowise GitHub releases page for migration notes — some major versions require manual migration steps.