SoFunction
Updated on 2025-04-04

Complete guide to installing DeepSeek on Windows (latest recommendation)

A complete guide to installing DeepSeek on Windows

  • DeepSeek runs locally without an internet connection.
  • Ollama and LM Studio are the key to installing and using the model.
  • Version 8B is ideal for computers with basic resources.
  • LM Studio provides a graphical interface for easy use.

DeepSeek It has become one of the most striking AI is its ability to run locally without requiring an internet connection. In this article, we will teach you how to install this AI on a Windows computer and how to make the most of it with the following tools: Orama y LM Studio, the options are attracting attention for their versatility and ease of use. Here you can find detailed information on all necessary steps, avoid complex technical issues and facilitate installation

This process also applies to other operating systems, such as macos o GNU/Linux, but we will focus on installing on Windows. There are many benefits to running DeepSeek on a local computer, such as having better control over your data and using AI offline at any time. Let's take a look step by step.

Step 1: Download and install Ollama

Ollama is an important tool for getting DeepSeek to run on your computer. The software allows you to run AI models locally in an accessible way. Please follow the following steps: 1. Add to the official page Orama (). 2. Download the program version compatible with your operating system. For Windows, this is an executable file (.exe). 3. Double-click the downloaded file to start the installation process. 4. Follow the on-screen instructions. You just need to click "Install" and wait a few minutes until the installation is complete.

Step 2: Download the DeepSeek R3NUMX model

Once Ollama is up and running, the next step is to download the AI ​​model you want to use. Here we will focus on the 8B version, which is very suitable for computers with not very high specifications. Open a command prompt in Windows. You can implement "cmd" in the search bar by: Enter the following command to download the model: ollama pull deepseek-r1:8b. This will start downloading the streamlined DeepSeek model. Wait patiently; this file takes up about 4.7 GB. During this process, do not close the window or close the computer. After the download is completed, the model will be ready for use. We'll go to the next step

Step 3: Run DeepSeek

After downloading the model, you can now start with DeepSeek. To run it, follow these instructions: 1. Open the command prompt again. 2. Type the following command: ollama run deepseek-r1:8b. 3. DeepSeek will start and be ready to answer your question. Enter any message in the terminal and press Enter to get a response. If you want to end the session, enter /bye You will return to the initial state of the command prompt. DeepSeek also has the following features Additional commands You can consult by writing a letter /?.

Alternative: Use LM Studio

If you prefer to experience the graphical interface without having to operate from the terminal, LM Studio is a great choice. The software allows DeepSeek models to be loaded visually and facilitate interaction.

  • Download LM Studio from the official website.
  • Follow the steps in the Installation Wizard to install.
  • Open the program and search for the DeepSeek R1 model in its model finder.
  • Download version 8B or your favorite version according to the features of your computer.
  • Load the model and start interacting with it through a chat-like visual interface.

This option does not require comandos, which is ideal for beginners or inexperienced users. If you value privacy, because your data will not be shared with third parties. In addition, its offline ability makes it a versatile tool for anyone from students to developers. With options like Ollama or LM Studio, you have the flexibility to choose the configuration that best suits your needs and knowledge.

This is the article about this complete guide to installing DeepSeek on Windows. For more related content on Windows installation, please search for my previous articles or continue browsing the related articles below. I hope everyone will support me in the future!