Deepseek R1: How to Run It Locally for Maximum Privacy

This blog post explains the key differences between using the open-source version of Deepseek R1 locally and the free version available through its app or website. It highlights the importance of data privacy, provides step-by-step guidance on setting up Deepseek R1 to run offline on your PC or laptop, and recommends tools like LMStudio and Ollama for non-technical and technical users. Perfect for those looking to securely use AI locally, the post also offers expert assistance for setup.

Jason Siow

2/21/20252 min read

Deepseek R1 has been making waves since its release, with many users exploring its capabilities. However, there’s a critical distinction between the open-source version of Deepseek and the free version available through its app or website. Understanding this difference is crucial to ensure your data remains private.

What is Deepseek R1?

Deepseek R1 is an open-source large language model (LLM) that offers powerful AI capabilities. While it is open-source, most users are introduced to it through its official app or website, similar to how ChatGPT was initially promoted. This free version allows users to interact with Deepseek, but it comes with a significant caveat: your data may be shared with Deepseek servers.

Why You Should Run Deepseek Locally

If you’re using Deepseek for office or business purposes, it’s essential to stop using the app or website immediately. Your data is being shared outside your security zone. Instead, you should download Deepseek and run it locally on your PC or laptop. This ensures that all your interactions, file uploads, and chats remain private and secure within your device.

How to Set Up Deepseek Locally

Setting up Deepseek locally requires some IT and AI knowledge, but it’s worth the effort for enhanced privacy. Here’s a step-by-step guide to get started:

1. Choose a program like Ollama or LMStudio to prepare your PC or laptop for running LLM models locally.

2. Install the program and check your device’s compatibility with the Deepseek R1 model.

3. Search for the Deepseek R1 model within the program and download it.

4. If you’re using LMStudio, enjoy a user-friendly chat interface similar to ChatGPT. Ensure the model is loaded by checking the model name section at the top.

5. For advanced users, Ollama offers a more technical setup, where you interact with the model via a terminal. This requires downloading the model from GitHub.

Which Option is Best for You?

For non-technical users, LMStudio is the recommended choice due to its simplicity and ease of use. Advanced users with IT knowledge may prefer Ollama for its flexibility and terminal-based interaction. Regardless of your choice, running Deepseek locally ensures complete privacy and control over your data.

Need Help Setting Up?

If setting up Deepseek locally seems overwhelming, don’t worry! I can assist you either on-site or remotely to configure your PC or laptop. The process takes just 1-2 hours, and I guarantee that your data will remain secure. My role is simply to set up your device for the LLM to run locally, leaving you in full control of your interactions.

Follow me and like this blog for more AI tips and tutorials! Check out my other posts to become an advanced AI user. I also have a detailed guide on hardware requirements for running LLMs locally—don’t miss it!