How to install DeepSeek locally for free
- 3rd Feb 2025
Table of Contents
When I started prototyping my online escape room app, I knew if I figured out how to install DeepSeek locally, it would mean faster responses, better control over data, and no dependency on external servers. If you’re working on an interactive project like mine, or just want your own private AI, here’s how to set it up on Windows.
Prerequisites: It’s good to have atleast 20gb storage and a powerful graphics card to make the most out of this tutorial.
Step 1: Get Ollama Up and Running

- Open your browser, search for Ollama, and download it.
- Run the installer, click Next a few times, and let it do its thing.
- If Ollama is running in your system tray, you’re good to go.

Step 2: Install DeepSeek locally
- Head to the Models section on Ollama’s site.
- Find DeepSeek R1, choose the version you want (7B is a solid choice), and copy the install command.

ollama run deepseek-r1:7b
3. Open PowerShell, paste the command, hit Enter, and let it install.

Now you can talk with DeepSeek!

You can choose to stop the tutorial here if you’re happy with a command-line interface, but if you want a more modern UI like ChatGPT read on to see how.
Step 3: Prepare to Upgrade to a Better Interface with Docker
- Install and download Docker.
- Restart your PC once installation is complete.
- Open Docker and create an account if needed.
Step 4: Set Up Open Web UI for a Cleaner Experience
- Open Powershell again and enter the following two commands:
docker pull ghcr.io/open-webui/open-webui:main
docker run -d -p 3000:8080 -v open-webui:/app/backend/data --name open-webui ghcr.io/open-webui/open-webui:main

Step 5: Start Using DeepSeek AI locally
- Open your browser, navigate to https://localhost:3000, and start chatting. It should have autodetected the DeepSeek installed previously.
- Be patient—running locally isn’t always as fast as cloud-based AI, but it keeps your data private.

Step 6: Uninstalling If Needed
- Open Control Panel, uninstall Docker and Ollama.
- Delete the Docker and Ollama folders from
C:\Users\YourUsername
. - Disable Windows Subsystem for Linux (WSL) in Windows Features, then restart.
Now You Have Local AI Power
Whether you’re crafting an AI-powered escape room app or simply exploring the depths of your creative projects,
DeepSeek offers a powerful tool right at your fingertips—completely independent from external servers. By setting
this up locally, you gain the freedom to experiment freely and maintain absolute control over your data.
So, go ahead and give it a try—what are you waiting for? Share this article with friends who might benefit from
learning about this incredible tool. Maybe even try using AI to create your own online escape room?