What's installed on your Langflow server?
- Langflow 1.13 with its visual editor, Playground, API and MCP server
- An admin account ready to go: the login is yonetici, Turkish for admin, and automatic login is off
- New accounts stay inactive until the admin activates them
- The default SQLite database
- A secret key generated on your server, which encrypts your model keys
- Usage reporting turned off
- HTTPS through Caddy; firewall with only ports 22, 80 and 443 open
What size do you need for Langflow?
Langflow's install guide lists a dual-core CPU and 2 GB of RAM as the minimum, and recommends a multi-core CPU with at least 4 GB. That's why the smallest size here is 4 GB. The model runs elsewhere, so the load comes from how many flows you have, how many run at once and the documents you process. As a rule of thumb, take 4 GB for a few flows, 8 GB for a team and 16 GB for heavy document work. There's no graphics card, so a local model inside a flow runs on the CPU and replies slow down.
| Use | Size | Price | Choose |
|---|---|---|---|
| Starter | 4 GB RAM · 2 vCore · 60 GB NVMe | $6.91/mo | |
| Medium | 8 GB RAM · 4 vCore · 120 GB NVMe | $13.63/mo | |
| Powerful | 16 GB RAM · 8 vCore · 200 GB NVMe | $25.18/mo |
Who is it for?
- Developers prototyping RAG and agent flows by drag and drop
- Anyone who wants to plug a flow into an app as an API or MCP server
- Teams who want to change components in Python
- People who'd rather keep flows and model keys on their own server
What happens after you order?
- Set the size and complete your payment.
- Your server starts and Langflow installs itself.
- Open your server from My servers in the panel. On the Overview tab, wait until the App row says it's ready, then click Open app.
- Log in with the details from the Login and Password rows. The login is yonetici, and there's no setup screen.
- Enter your model keys under Global Variables in the settings.
- Build your first flow, try it in the Playground, then call it through the API.
With every server
- Daily backups included in the price
- Unlimited traffic
- Full root access
- Scale RAM and CPU up or down later, and grow the disk
- Location: Istanbul · IST-1.
Good to know
- Your model providers' usage charges aren't included in the price. You enter the keys and settle the bill with each provider.
- Anyone who can edit flows can run Python code on the server. Only give accounts to people you trust.
- The visual editor has no user management: since Langflow 1.12, the admin adds and activates users through the Users API.
- Updates are your responsibility. Export your flows as JSON before upgrading: this install uses the default SQLite database, and data can be lost when the container is recreated.
- Global variables are encrypted with the secret key. If you lose the key in /opt/seturack/langflow/.env, your saved model keys can't be decrypted.
Langflow questions and answers
How much RAM does Langflow need?
Langflow's documentation gives 2 GB as the minimum and recommends at least 4 GB, so our smallest size is 4 GB. As a rule of thumb, look at 8 GB for a team and 16 GB for heavy document work. You scale up from your panel, and your data and IP address stay the same.
How do I install Langflow?
You don't have to: Langflow installs itself on your server after payment, and the site address, login and password appear in your panel. Root access is yours as well, so you can connect over SSH and change the settings in /opt/seturack/langflow whenever you like.
Does Langflow need a GPU?
No. Langflow runs on the CPU and flows usually call a hosted model. Seturack servers have no graphics card. If you run a local model inside a flow it runs on the CPU, which works for short tasks, but replies come noticeably slower than on a GPU.
How do I call a Langflow flow as an API?
Create an API key in Langflow, use the flow's API path with your Site address, and send the key in the x-api-key header. You can also publish flows as an MCP server and call them as tools from MCP clients.
Which models can I use in Langflow?
Langflow supports the major model providers and vector databases. Put a provider's key, such as OpenAI's, under Global Variables in the settings and select it in your flow. The install doesn't download a model, and usage is billed by the provider.
How do I back up my Langflow flows?
Your server is backed up every night, the last 7 days are kept, and that's included in the price. You can also export any flow or a whole project as JSON, and use the same file to bring it into another Langflow.