Advertising disclosure: We may earn a commission when you buy through links on this page. It does not change the price you pay or the order of the list.
Open WebUI: a private ChatGPT on your PC or your own server
Open WebUI is a free, self-hosted chat interface that runs in your browser and talks to local models through Ollama or to any OpenAI-compatible API. Most people install it with Docker on their own PC or a VPS; it adds chat history, document uploads and user accounts, while the widget below shows which models your hardware can serve.
Which models fit
Runs on it
- DeepSeek-R1-Distill-Qwen 32B · Q6~40 tokens/s
- Qwen3 32B · Q6~40 tokens/s
- Qwen3 30B-A3B (MoE) · Q6~150+ tokens/s
- Gemma 3 27B · Q8~37 tokens/s
- Mistral Small 3.2 24B · Q8~42 tokens/s
- Magistral Small 1.2 24B · Q8~42 tokens/s
- Mistral Small 24B · Q8~43 tokens/s
- gpt-oss 20B · MXFP4~150+ tokens/s
Too big for it
- Llama 3.3 70BRent a GPU →
- DeepSeek-R1-Distill-Llama 70BRent a GPU →
- Qwen2.5 72BRent a GPU →
Best quality that fits, 8,192 tokens of context, one user. Speed is an estimate from memory bandwidth (generation reads every active weight once per token); real numbers depend on the program and settings.
On your computer or on a server?
On your own computer
Full privacy and no subscription; the model size is limited by your memory.
Which hardware →On a VPS
Reachable from anywhere and shareable with a team; small models run on CPU.
Ollama on a VPS →What Open WebUI gives you
Open WebUI is the front end, not the model. It looks much like ChatGPT: a sidebar of conversations, a model picker, file and image uploads, and answers that can cite the documents you attached. Behind it you plug in a model source, most often Ollama on the same machine, but also LM Studio’s server, vLLM or a paid API that speaks the OpenAI format.
That split is useful. You can keep the same interface while you swap models, mix a local model for private work with a cloud model for harder questions, and decide yourself where each conversation goes.
Beyond plain chat, it can search the documents you upload and quote them in answers, keep reusable prompts, run web searches if you enable them, and talk to image generators. Each extra feature is optional, so you can keep a setup that never leaves your machine, or add connections one by one when you need them.
Open WebUI with Docker, on a PC or a VPS
Docker is the recommended install: one command pulls the image, keeps your data in a volume and survives updates. On your own PC that is all you need, with Ollama running alongside. On a VPS, the interface is reachable from anywhere, so put it behind HTTPS with a reverse proxy such as Caddy or Nginx and keep sign-ups closed.
Remember what the server can run. An ordinary VPS has no GPU, so pair it with small models, or with a remote model source: a rented cloud GPU running Ollama, or an API.
Using it as a team
The first account you create becomes the administrator. You can then invite colleagues, approve new users, decide which models each group can use, and share prompts and knowledge bases. For a small company this is the quickest way to give everyone a private chat assistant without sending documents to a third-party chat service. Size the hardware for how many people will ask questions at the same time, not for the total number of accounts.
- Turn off open registration once your team is in
- Back up the Docker volume that holds chats and settings
- Update the image regularly: it is an internet-facing app
Install and run it in 5 steps
- Install Docker Desktop on your PC, or Docker Engine on your VPS, and make sure Ollama is installed if you use local models.
- Run the docker run command from the Open WebUI docs, which maps a local port and creates a volume for your data.
- Open the address in your browser and create the first account; it becomes the administrator.
- In the admin settings, check that Ollama is connected, or add an OpenAI-compatible API address and key.
- Pick a model that fits your hardware in the widget above, pull it, and start chatting or invite your team.
Questions people ask
What is Open WebUI?
Open WebUI is a free, self-hosted chat interface for AI models that runs in your browser. It connects to Ollama for local models or to OpenAI-compatible APIs, and supports several users with their own accounts.
How do I install Open WebUI with Docker?
Install Docker, then run the docker run command from the official Open WebUI documentation. It starts the container with a volume for your data; open the mapped port in your browser and create the admin account.
Does Open WebUI need Ollama?
No. Ollama is the most common pairing for local models, but Open WebUI also works with any OpenAI-compatible API, including LM Studio’s server, vLLM and paid cloud APIs.
Can I run Open WebUI on a VPS?
Yes, and it is a good way to reach your chat from anywhere. Use HTTPS, close public sign-ups, and remember a normal VPS has no GPU, so use small models or a remote model source.
More on local AI
Too slow? Rent a GPU from $0.33/hSee plan →