Self-Hosted AI / Agent builder

Flowise

Drag-and-drop LLM app and agent builder - no code required to wire up a RAG pipeline.

What is Flowise?

Flowise is a visual builder for LangChain-style chains and agents - instead of writing Python to wire a prompt to a model to a vector store, you drag nodes onto a canvas and connect them.

It's the same category as Langflow and Dify: aimed at people who want to build something like 'a chatbot that answers questions from our internal docs' without standing up a full development environment first.

Every flow you build can be exposed as its own API endpoint, so it plugs into an existing app or website once you're happy with it.

How DeployOS deploys it

  1. 1DeployOS deploys the official flowiseai/flowise image with a persistent volume at /root/.flowise, so saved flows, credentials, and uploaded documents survive restarts.
  2. 2It's published through your reverse proxy with automatic HTTPS, same as any other DeployOS app - no separate certificate or domain setup.
  3. 3Point it at a self-hosted model (Ollama) or a hosted API key from inside its UI, and start connecting nodes.
Imageflowiseai/flowise:latest
Port3000
Persistent volume/root/.flowise

Pairs well with

Common questions

Do I need to know LangChain to use this?

No - that's the point of the visual builder. Familiarity with LangChain's concepts (chains, retrievers, agents) helps you use the nodes faster, but you don't write any of that code directly.

Can it use my own documents for answers?

Yes - connect a document loader node to a vector store (Qdrant or Chroma, both one-click DeployOS templates) and a retrieval-augmented-generation flow is a few connected nodes, not a custom pipeline.

Deploy Flowise on your own server in one click.

Free download, no account needed. AI templates are on paid plans.

Download DeployOS