Running DeepSeek Locally: A Step-by-Step Guide
Javier Machin

Jan 11, 2025

Running DeepSeek Locally: A Step-by-Step Guide

Learn how to install DeepSeek locally in just a few simple steps. Secure, private, and free.

engineeringtutorialLLMAIDeepSeek

Hey there, AI explorers! 👋 Today, we're diving into how to run DeepSeek locally using LM Studio, which is free and doesn't require registration. With just a few simple steps, you can harness the power of AI on your own machine, securely, privately and without any usage limits. Let's get started!

Step 1: Download LM Studio

First things first, head over to LM Studio's download page. Select your operating system and click on the download button. This is your gateway to a local AI experience.

Step 2: Install LM Studio

After downloading, run the installer and follow the on-screen instructions. Once installed, launch LM Studio to begin exploring its features.

Step 3: Discovering DeepSeek Models

Inside LM Studio, locate the sidebar on the left and click on the Discover tab (look for the purple icon). Here, you can search for models that suit your hardware.

If you're unsure about your computer's capabilities, start by searching for DeepSeek Qwen 7b. For those with a modern Mac or a powerful gaming PC, opt for DeepSeek Qwen 14b for an enhanced experience.


Step 4: Download and Load Your Model

Once you've chosen your model, click the download button located at the bottom right of the window. Keep in mind, the download might take some time as files range from 4.5GB to 9GB.

After the download completes, navigate to the Chat tab in the sidebar. At the top, you'll find a prompt saying "select a model to load". Choose the model you just downloaded.


A new window will appear, offering configuration options. If you're new to this, simply leave the settings as they are and click on Load Model.

Step 5: Start Chatting

Within seconds, your model will be up and running. Now, you can chat with DeepSeek just like you would with any AI chat platform like ChatGPT, Claude, or the web version of DeepSeek, but with the advantage of local privacy and no usage restrictions. The downside is that these models are not as performant as their full size commercial counterpart, but still quite capable!


Conversations are stored only on your computer, giving you full control. You can easily delete previous chats or start new conversations from the sidebar.


LM Studio is quite powerful and if you are a developer it can be used to provide an API to interact with the models.

Enjoy your journey with DeepSeek and LM Studio, and embrace the freedom of a self-hosted AI experience! 🚀

Service Workers vs. Web Workers: What's the Difference?

Discover their superpowers and when to use them

Sep 25, 2020Read more