Back to blog

A Guide to Local Deployment of DeepSeek-R1: Ollama and Open-WebUI Setup

Artificial Intelligence has made massive leaps forward, and one of the most exciting trends is the ability to run powerful models right on your local hardware. DeepSeek-R1 is a state-of-the-art reasoning model that delivers impressive results. By deploying it locally, you gain complete privacy, zero latency charges, and offline availability.

In this guide, we will walk you through setting up Ollama, downloading DeepSeek-R1, and connecting it to Open-WebUI for a beautiful, ChatGPT-like chat interface.

Prerequisites

Before starting, make sure your system has a dedicated GPU (NVIDIA, AMD, or Apple Silicon) for a smooth experience. While CPU inference is possible, it can be slow for larger model versions.

Step 1: Install Ollama

Ollama is a lightweight tool that makes running large language models locally simple and straightforward.

  1. Go to the official Ollama website and download the installer for your operating system (Windows, macOS, or Linux).
  2. Run the installer and follow the instructions.
  3. Open your terminal or Command Prompt and verify the installation by typing:
ollama --version

Step 2: Download and Run DeepSeek-R1

Ollama provides DeepSeek-R1 in different sizes (parameters). The smaller models (such as 1.5B, 7B, or 8B parameters) run extremely well on typical consumer laptops.

To download and run the 7B parameter version of DeepSeek-R1, run the following command in your terminal:

ollama run deepseek-r1:7b

Once the download finishes, you can start typing prompts directly in the terminal to chat with the model. Type /exit to close the prompt.

Step 3: Install Open-WebUI

While typing in the terminal works, a visual web interface is much more user-friendly. Open-WebUI is a feature-rich chat interface that connects directly to Ollama.

The easiest way to run Open-WebUI is by using Docker. Make sure Docker is installed and running on your computer.

Open your terminal and run the following command to download and start Open-WebUI:

docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main

This command runs Open-WebUI in the background and exposes it on port 3000.

Step 4: Access Your Private AI

Once the Docker container starts:

  1. Open your web browser and navigate to http://localhost:3000.
  2. Create your administrator account (this account is saved locally on your machine).
  3. In the top-left dropdown menu, select deepseek-r1:7b as your active model.
  4. Start chatting!

By running DeepSeek-R1 locally with Ollama and Open-WebUI, you now have a fully functional, secure, and private AI environment. You can explore other model sizes or customize your chat settings directly in the web dashboard.