Langflow on your own server for AI agents and RAG flows
Design agents, retrieval flows and multi-agent systems on a canvas, test them in the playground and serve each flow as an API or an MCP server. Run it with PostgreSQL on a HyperDC server, preinstalled as an app option or set up with our guide.
- Drag-and-drop flows for agents, RAG and tools
- Every project can act as an MCP server
- Custom components in Python when you need them
- Runs from 2 cores and 2 GB of RAM, 4 GB recommended
- Stack
- Python 3.10–3.14, Docker image
- Default portLangflow’s documentation warns never to expose its port to the internet without security measures: put it behind a reverse proxy with HTTPS and authentication.
- 7860
- Minimum
- 2 cores, 2 GB RAM (4 GB recommended)
- Your data
- SQLite, or PostgreSQL in production
- Official docs
- docs.langflow.org
Build with
- Agents
- RAG
- MCP
- Playground
- Python components
- LangSmith
- Langfuse
- Ollama
Facts from the project’s official website, documentation and repository, checked in October 2026.
Plans are being prepared
We are preparing ready-to-use plans for Langflow. Tell us how you will use it and how many users you expect, and we will reply with a server that fits. You can also start today on a Linux VPS and install it with our guide.
Which server size fits?
Starting points for vCPU, memory and disk. Grow the server when your data and users grow.
| Feature |
Builder
One person designing flows
|
Recommended Team
Shared flows served as APIs
|
Production
Busy APIs and large vector stores
|
|---|---|---|---|
| vCPUVirtual processor cores of the server. | 2 | 4 | 8 |
| MemoryMemory for the app, its database and the operating system. | 2–4 GB | 8 GB | 16 GB |
| DiskNVMe or SSD storage for the app, its data and local backups. | 20 GB | 50 GB | 100 GB+ |
| Database | SQLite | PostgreSQL | PostgreSQL, vector database |
| Server type | Linux VPS | Linux VPS | VPS or VDS |
-
Builder
One person designing flows
- vCPUVirtual processor cores of the server.
- 2
- MemoryMemory for the app, its database and the operating system.
- 2–4 GB
- DiskNVMe or SSD storage for the app, its data and local backups.
- 20 GB
- Database
- SQLite
- Server type
- Linux VPS
-
Recommended
Team
Shared flows served as APIs
- vCPUVirtual processor cores of the server.
- 4
- MemoryMemory for the app, its database and the operating system.
- 8 GB
- DiskNVMe or SSD storage for the app, its data and local backups.
- 50 GB
- Database
- PostgreSQL
- Server type
- Linux VPS
-
Production
Busy APIs and large vector stores
- vCPUVirtual processor cores of the server.
- 8
- MemoryMemory for the app, its database and the operating system.
- 16 GB
- DiskNVMe or SSD storage for the app, its data and local backups.
- 100 GB+
- Database
- PostgreSQL, vector database
- Server type
- VPS or VDS
Langflow’s documentation gives a dual-core CPU and 2 GB of RAM as the minimum and at least 4 GB as recommended. Local models and large embedding jobs need their own resources.
What you can build with Langflow
Prototype in the browser, then serve the same flow to your apps.
Agents with tools
Give agents tools, memory and other agents to call, and watch every step in the playground.
Retrieval over your data
Load documents into a vector store and build question-answering flows with the model of your choice.
APIs and MCP
Call any flow over the API, or expose a project as an MCP server for Claude, Cursor and other MCP clients.
Your data stays yours
Prompts, files and databases stay on a server you control, in the location you choose, instead of a shared SaaS account.
Full root access
Install what the app needs, change any setting and run more services next to it. Nothing is locked behind a panel.
App option or step-by-step guide
Order the server with the app installed as an option, or set it up yourself on a clean Linux server with our guide.
Near your users
Choose a data center in the United States, Europe or Asia. The order form estimates the latency from where you are to each location.
Grow without starting over
Start on a VPS, then move to a bigger plan, a VDS with NVMe storage or a dedicated server when the workload grows.
From order to first login
Order the app preinstalled on your server, or install it yourself with our guide.
-
Pick the server
Choose a size from the table above and the data center closest to the people who will use the app.
-
Add Langflow
Select Langflow as an app option when you order, or install it on a clean Ubuntu or Debian server with our guide.
-
Point a domain and enable HTTPS
Create a DNS record such as app.example.com for the server and put a reverse proxy with a free Let’s Encrypt certificate in front of the app.
-
Secure the superuser
Set LANGFLOW_SUPERUSER, a strong password and your own LANGFLOW_SECRET_KEY, turn automatic login off, then sign in.
Step-by-step setup guides
Install, secure and update the app with our guides, written for current Ubuntu and Debian releases.
-
How to install Langflow with Docker Compose and PostgreSQL
Run the official Langflow image with PostgreSQL, turn off auto-login with a superuser and your own secret key, publish it over HTTPS with Caddy, connect Ollama and keep code execution private.
35 min Intermediate
Related solutions
Frequently asked questions
How much server does Langflow need?
Langflow’s installation guide lists a dual-core CPU and 2 GB of RAM as the minimum and a multi-core CPU with at least 4 GB as recommended. Add memory for many concurrent API calls, and run local models on their own server.
Is Langflow safe to expose to the internet?
Only with protection. The documentation warns never to expose Langflow’s port directly. Turn off automatic login (pip installs enable it by default; the official Docker images do not), set a superuser with a strong password and your own secret key, and publish it through a reverse proxy with HTTPS.
Which database should I use?
Langflow uses SQLite when no database URL is set, which is fine for one builder. For a team or production APIs, use PostgreSQL through LANGFLOW_DATABASE_URL; the official Docker Compose example includes it.
Can Langflow work with local models?
Yes. Components connect to Ollama and other OpenAI-compatible servers as well as hosted providers. Point them at an Ollama or vLLM server on a GPU server for larger models.
What is the MCP server feature?
Each Langflow project can act as an MCP server, so MCP clients such as Claude Desktop or Cursor can call your flows as tools. Langflow can also use tools from other MCP servers inside a flow.
How do I call a flow from my app?
Every flow has an API endpoint. Create an API key in Langflow, then send requests from your backend with the key; the API panel in the editor shows ready-made examples.
Why is there no Flowise page?
Flowise’s maintainers archived the project in August 2026 and ended its support, so we do not offer new Flowise installations. Langflow and Dify cover the same visual agent and RAG use cases and are actively developed; our team can help you move existing flows.
Ready to build your first flow?
Tell us the app, how many people will use it and where they are, and we will suggest a server for it.