How to install and use DeepSeek on your computer: Complete guide

Last update: January 28, 2025
Author TecnoDigital
  • DeepSeek is an open source artificial intelligence ideal for complex tasks.
  • Installation is done through Ollama, simplifying the process on any operating system.
  • It can be integrated into local projects or via API for large-scale applications.

deepseek install deepseek If you are looking for how to install DeepSeek on your computer, you've come to the right place. This guide will take you step by step to configure this powerful Artificial Intelligence on your system, whether you use Windows, MacOS o GNU / Linux. The best thing of all is that you can run it locally, which guarantees privacy or with a continuous access, even without connection to Internet.

DeepSeek is a tool of Artificial Intelligence open source code that stands out for its deep reasoning capabilities. This feature makes it ideal for solving complex problems such as advanced logic, mathematics or even programming tasks. Do you want to avoid high costs and keep your data safe? Here you will learn how to do it.

What do you need to start?

Before you start, it's important to prepare your computer with some basic requirements. Here are some things you need to consider depending on the operating system you use:

  • Windows: You just need access to the command prompt and a simple program like Don't.
  • macOS: Make sure you familiarize yourself with the terminal and download Don't to manage AI.
  • Linux: Use the terminal to configure and install tools like Docker if you want a graphical interface.

Ollama's role in the installation

Don't is the main intermediary to execute DeepSeek on your computer. It acts as a platform to download and manage models of IA locally. Why is this important? Basically, it simplifies the process and ensures that the model works in your environment without any further complications.

  Voice mode in ChatGPT: how to use it, what it offers, and what its limits are

To install Don't, visit their official website (ollama.com) and download the program compatible with your operating system. The installer will guide you through the process, which consists of simply accepting the conditions and clicking “Install”.

DeepSeek Installation Step by Step

Once you Don't is installed, it's time to configure DeepSeekThe process varies slightly depending on the resources available on your computer, but here are the general steps:

  • Open Terminal or Command Prompt. In Windows, search for “cmd” in the start menu. In MacOS y GNU / Linux, access directly from the respective applications or search engines.
  • Run the command to install DeepSeek depending on the version you prefer. For example:
    • ollama pull deepseek-r1:1.5b for the lighter version.
    • ollama pull deepseek-r1 for a more complete model, but one that requires greater capacity.
    • ollama pull deepseek-r1:70b for advanced tasks, if you have high-end hardware resources.
  • Once downloaded, launch the program by typing ollama run deepseek-r1.

Using DeepSeek in your language

DeepSeek It has a very interesting advantage: it can respond in Spanish if you configure it that way from the start. Make sure to make your initial queries or "prompts" in this language so that the Artificial Intelligence Please continue to use the same one. If you ever respond in English, add to the command “Respond in Spanish” and it will adjust its behavior.

The model will process all your queries from the terminal, even presenting the stages of reasoning before arriving at a final answer. This transparency is one of the most fascinating features of DeepSeek.

Advanced Options: Docker and GUI

If you prefer to work with a visual interface Instead of the console, you can set up a friendly UI using DockerThis step is optional and is intended for those with intermediate knowledge of containers and local networks.

  1. Download and install Docker from its official website (docker.com).
  2. Open the command line and run the following code to download a compatible graphical interface: 
    docker run -d -p 8080:8080 \ 
    -v open-webui:/app/backend/data \ 
    -e OLLAMA_BASE_URL=http://127.0.0.1:11434 \ 
    --name open-webui \ 
    ghcr.io/open-webui/open-webui:main
  3. Access the interface in your browser by entering http://localhost:8080.
  4. Choose the model of DeepSeek that you prefer and start using it visually.
  Artificial Intelligence Engineering: 10 Keys to Mastering the Technological Future

Integration with projects

DeepSeek It is not only useful as an assistant in your terminal. You can also integrate it into development projects, either locally or through the cloud with APIs.

For local use, you can configure an endpoint that works with popular libraries such as OpenAI en PythonWith a simple adaptation, your programs will be able to make requests directly to the model that is running on your machine.

If you need a greater reach, DeepSeek also offers one API official that you can use for large scale projects. The API It is similar to that of OpenAI, which makes it very versatile.

DeepSeek It is an outstanding tool for those looking for a Artificial Intelligence Powerful and customizable. With its ability to run locally, you avoid high costs and retain control over your data. Whether you choose to use it from the console, through a graphical interface, or integrate it into your projects, you will have a flexible and economical solution for advanced reasoning tasks.