Must have
Installation requirements
- PostgreSQL — Recommended for production; SQLite is the default
- Model providers — API keys or endpoints for the models your flows call
Application hosting
Hosting Langflow
Langflow is a visual builder for AI agents and retrieval workflows that you can also serve as APIs. It calls models from providers or your own servers; the Langflow server itself needs CPU, memory and a PostgreSQL database for production.
Requirements last checked against the project's documentation.
A sensible first server for Langflow is 2 vCPU, 4 GB RAM and 20 GB of storage. The cheapest listed plan that fits is OVHcloud VPS-1 (US) at $4.54/month.
Understand the setup
Find the software and services needed to run the application.
See requirementsBudget your resources
Plan memory, storage and CPU for the way you will use it.
Open resource plannerBuild your shortlist
Compare plans for the starting size, then review the provider and billing terms.
See plansRecommended
Planning guidance
- Use an external PostgreSQL database for production instead of the default SQLite.
Choosing a server for Langflow
Flow complexity, concurrent runs and payload size drive resource use. Model inference happens at the providers or model servers a flow calls.
Where to start
A sensible first server for Langflow
- vCPU
- 2
- RAM
- 4 GB
- Disk
- 20 GB
Langflow's production guide sets a minimum of 1 CPU and 2 GiB of RAM per runtime instance. We add room for PostgreSQL on the same server; that and the disk size are our estimates.
Langflow production best practices · Size it for your workload
Our shortlist
Cheapest plans that meet this size
-
$4.54/mo list price
-
$5.00/mo list price
-
$10.00/mo list price
-
$14.99/mo list price
-
$15.99/mo list price
-
$17.00/mo list price
-
$18.88/mo list price
-
$18.99/mo list price
One cheapest plan per provider, based on each provider's standard list price. Check the plan page for introductory and renewal pricing before you buy.
Deploy with
Supported paths
The host must provide
Runtime
- PostgreSQL
- Recommended for production; SQLite is the default
- Model providers
- API keys or endpoints for the models your flows call
Storage
Persistent data
By default the SQLite database lives inside the container and is lost when it is recreated. Mount a volume at the configured LANGFLOW_CONFIG_DIR, such as /app/langflow, or point LANGFLOW_DATABASE_URL at PostgreSQL.
Step by step
Plan a Langflow deployment
-
Run a pinned image with persistent data
Use a pinned langflowai/langflow version tag and mount a volume for the config directory, or the flows and settings disappear with the container. The interface listens on port 7860.
Official instructions -
Move to PostgreSQL for production
Point LANGFLOW_DATABASE_URL at PostgreSQL with SSL, and set a fixed LANGFLOW_SECRET_KEY in the environment; an auto-generated key is not enough for production.
Official instructions -
Enable authentication
Start the server with authentication enabled before exposing the editor or flow APIs.
Official instructions
Security
Before you go public
- Enable authentication before the editor or flow APIs are reachable from the internet.
- Set LANGFLOW_SECRET_KEY as a fixed secret and connect to PostgreSQL over SSL.
- Serve Langflow behind HTTPS and keep the database port private.
Cost
Beyond the server price
- The server is modest; model calls made by your flows are billed by their providers.
- Production adds a PostgreSQL database and, for high availability, several Langflow instances.
Want the full monthly bill? Use the hosting cost calculator. Add storage, backups and other costs to your estimate.