$ ollama pull llama3.2:3b
$ ollama pull llama3.2:3b
$ ollama pull llama3.2:3b
$ ollama run llama3.2:3b
$ ollama run llama3.2:3b
$ ollama run llama3.2:3b
-weight: 500;">docker
$ -weight: 500;">docker run -d -p 3000:8080 --name open-webui ghcr.io/open-webui/open-webui:ollama
-weight: 500;">docker
$ -weight: 500;">docker run -d -p 3000:8080 --name open-webui ghcr.io/open-webui/open-webui:ollama
-weight: 500;">docker
$ -weight: 500;">docker run -d -p 3000:8080 --name open-webui ghcr.io/open-webui/open-webui:ollama - What Ollama and Open WebUI are (and why they work so well together)
- How to set them up locally (no cloud required)
- How to deploy them on a server for 24/7 access from anywhere
- What you can actually build with your own private AI - Ollama is the engine – it runs the models.
- Open WebUI is the dashboard – it gives you a clean interface to talk to those models.
- Together, they create a private, fully self-hosted ChatGPT alternative that you control completely. Your conversations never leave your hardware. There are no usage caps, no subscription fees, and no data being sold or trained on. - Docker installed (Docker Desktop for Windows/Mac, or Docker Engine for Linux)
- At least 8GB of RAM (16GB is better for larger models)
- 10GB+ free disk space (models are 4–8GB each) - Downloads the Open WebUI container with Ollama pre-integrated
- Maps port 3000 on your computer to port 8080 inside the container
- Starts the container in the background - Go to Admin Panel → Settings → Document Settings
- Enable the RAG pipeline
- Choose a vector database (Chroma is the simplest to -weight: 500;">start with)
- Upload a file using the paperclip icon in the chat - Visit the Railway template page
- Click Deploy Now
- Railway provisions both services, attaches storage volumes, and gives you a public URL within minutes
- Set up your admin account when you first visit the URL