Open Source
ChatGPT-Style UI
Own Your Data

Deploy Open WebUI on CometVPS

A polished, ChatGPT-style web interface for any LLM you want to run — Ollama, OpenAI, Anthropic, or your own endpoints. Multi-user, RAG-ready, and entirely self-hosted.

Get Started with Open WebUILearn More About Open WebUI

What is Open WebUI?

Open WebUI is the most popular open-source front-end for self-hosted AI. It gives you a ChatGPT-style chat interface, document upload and RAG, role-based user management, and connectors for any OpenAI-compatible API — all in a single Docker container you control.

Private by Default

Conversations, documents, and users live in your database. Nothing leaves the server unless you point it at a remote model API.

Bring Any Model

Local Ollama models, OpenAI, Anthropic, Groq, OpenRouter, or anything proxied through LiteLLM — all in one chat dropdown.

Built for Teams

Roles, per-user model permissions, shared chats, and an admin panel make it easy to roll out as an internal AI tool for your org.

Key Features

A complete chat platform you fully control

ChatGPT-Style Interface

Familiar conversation UI with markdown rendering, code highlighting, conversation history, and streaming responses. Anyone who's used ChatGPT will feel at home.

Connect Any Backend

Works with Ollama, OpenAI, Anthropic, Groq, OpenRouter, LiteLLM, and any OpenAI-compatible endpoint. Switch models per conversation from a dropdown.

RAG & Document Chat

Upload PDFs, docs, and websites, then chat with them. Built-in vector store and embeddings — no separate retrieval pipeline to wire up.

Multi-User & Roles

Invite teammates with admin, user, or pending roles. Per-user model permissions, conversation isolation, and a built-in user management panel.

Installation Guide

Get Open WebUI running on your CometVPS server in 5 simple steps

Security Tip

The first user to sign up becomes the admin — register your own account immediately, then disable open sign-ups in the admin panel so strangers can't claim accounts on your instance. Always run behind HTTPS, and store third-party API keys (OpenAI, Anthropic) in environment variables rather than in the UI when possible.

Recommended VPS Plans

Choose the right server based on whether you'll run local models alongside the front-end

Solo / Light Use

Core VPS - Basic

$14/mo

2 vCPU Cores, 4GB RAM, 60GB NVMe

Open WebUI is lightweight on its own — this fits the front-end plus a small SQLite database when your LLM lives on a separate Ollama or OpenAI endpoint.

Order NowView All Plans
Recommended
Front-End + Local Ollama

Supernova VPS - Flare

$58/mo

6 AMD Ryzen Cores, 16GB DDR5, 200GB NVMe

Run Open WebUI and Ollama on the same box for a complete self-hosted ChatGPT — fast Ryzen cores keep 7B–13B models snappy.

Order NowView All Plans
Team Deployment

AstroMetal - AM-R7950X3D

$199/mo

16 Cores, 128GB RAM, 1.92TB NVMe

Multi-user team chat with role-based access, larger local models, and plenty of room for RAG document storage and embedding databases.

Order NowView All Plans

Why Deploy on CometVPS?

Get the best performance and reliability for your Open WebUI instance

Full Privacy

All conversations and uploaded documents stay on your server.

NVMe SSD Storage

Fast disk I/O for chat history, embeddings, and uploaded knowledge bases.

10Gbps Network

Premium network connectivity for fast streaming responses to remote users.

24/7 Expert Support

Our team is here to help with server issues around the clock.

Ready to Deploy Open WebUI?

Get a self-hosted ChatGPT for you and your team in minutes. Pair it with Ollama on the same server for a fully private AI stack.

Launch Your ServerAdd Ollama