How much does it cost to host Open WebUI?

$29.92 a month at most for the app, its 10 GB volume and its attached PostgreSQL database, billed by the running minute. Model usage is billed separately by the model provider you connect.

Do I need Ollama or a GPU to run Open WebUI?

No. Open WebUI can use any OpenAI-compatible API instead of local models. Add the endpoint URL and API key in Admin Settings, then choose a model.

How do I connect Open WebUI to the Varity AI Gateway?

Create an AI Gateway API key, then add an OpenAI-compatible connection in Admin Settings with https://ai.varity.app/v1 as the URL. Save, pick a model and send a test prompt.

Can my whole team use one Open WebUI?

Yes. The first account becomes the administrator, who invites users and controls who can sign up. Open WebUI's license requires keeping its branding once a deployment serves more than 50 users in 30 days.

Open WebUI

Open WebUI is a ChatGPT-style chat for your team, connected to the models you choose. Accounts, chats and uploaded files stay in your own deployment.

Open WebUI hosting on Varity costs $29.92 a month at most, with PostgreSQL and a 10 GB volume included. No request, ingress or egress fees.

$29.92/mo max

Deploy Open WebUI

$5 starter credit for apps, payment method required, 14 days

What you get

Deploying the template sets up every part below in your account, on its own HTTPS address

Services & storage

3 parts

Open WebUIChat UI2 vCPU · 4 GB RAM
PostgresRelational databasePersistent storage, 2 GB
10 GB volumeMounted in your appKept across restarts and redeploys

Why run Open WebUI on Varity

Private models in one connection
Add the Varity AI Gateway as an OpenAI-compatible connection. Pick a Private model and your prompts and replies aren't stored or used for training.
Chats and files that persist
PostgreSQL holds users and conversations; the 10 GB volume keeps uploads. Both are inside the $29.92 monthly maximum.
The interface has a ceiling
The app costs $29.92 a month at most, billed by the running minute. No request, ingress or egress fees. Model usage is billed separately from this app.

What Open WebUI runs on

Monthly maximum for Open WebUI, by part
Included	Monthly max
Open WebUI appScale plan: 2 vCPU · 4 GB RAM · 25 GB temporary storage	$28
App storage10 GB persistent volume, kept across restarts	$1.60
PostgreSQLAttached database, 2 GB storage	$0.32
Monthly maximum	$29.92

How app hosting is billed

Best for: Giving a team one browser interface for compatible model endpoints and conversation workflows
Billing: By the running minute, never above the maximum
Traffic: No app traffic fees
Category: AI agents
Built with: Python

Set up Open WebUI

01
In the Portal, choose Open WebUI from Templates and deploy it
02
Create the administrator account, then add an OpenAI-compatible connection in Admin Settings
03
Enter the endpoint URL and API key, save, and select a model
04
Send a test prompt and check the response before inviting others

Open WebUI setup guide Deploy your own code instead Need a model endpoint? AI Gateway

Open WebUI hosting questions

How much does it cost to host Open WebUI?: $29.92 a month at most for the app, its 10 GB volume and its attached PostgreSQL database, billed by the running minute. Model usage is billed separately by the model provider you connect.
Do I need Ollama or a GPU to run Open WebUI?: No. Open WebUI can use any OpenAI-compatible API instead of local models. Add the endpoint URL and API key in Admin Settings, then choose a model.
How do I connect Open WebUI to the Varity AI Gateway?: Create an AI Gateway API key, then add an OpenAI-compatible connection in Admin Settings with https://ai.varity.app/v1 as the URL. Save, pick a model and send a test prompt.
Can my whole team use one Open WebUI?: Yes. The first account becomes the administrator, who invites users and controls who can sign up. Open WebUI's license requires keeping its branding once a deployment serves more than 50 users in 30 days.

Keep browsing

Browse all 27 templates