# DeepSeek: Your Local AI Powerhouse with Ollama

The world of Large Language Models (LLMs) is rapidly changing, and running them locally is becoming increasingly appealing. DeepSeek, a powerful open-source LLM, combined with Ollama, makes this process surprisingly simple. This guide will walk you through setting up DeepSeek on your own machine.

![upload image deepseek](https://imgs.search.brave.com/cRcYOuQIrEUlridnua3wS9euiPdVWcVR9EE_0mJ5rXQ/rs:fit:860:0:0:0/g:ce/aHR0cHM6Ly93ZWJ1/c3VwbG9hZC5hcG93/ZXJzb2Z0LmluZm8v/Z2l0bWluZC93cC1j/b250ZW50L3VwbG9h/ZHMvMjAyNS8wMi9p/bWFnZS11cGxvYWQt/ZGVlcHNlZWsuanBn/LndlYnA align="left")

**DeepSeek: A Powerful LLM**

DeepSeek stands out for its strong coding abilities and general language understanding. It's open-source, allowing for transparency and community contributions, and importantly, it can be run locally, it’s also the first AI that took an AI’s Job xD(ChatGPT).

**Ollama: Simplifying Local LLM Use**

Ollama acts as a container and runtime for LLMs, making local deployment straightforward. It handles dependencies and configurations, allowing you to focus on using the model, Docker can also be used :)

**Benchmark below:**

![deepseek r1 benchmark](https://imgs.search.brave.com/ULV2HenLnwBqV9FkAKp6Empc6df9V9JVTrjgYZqqsRA/rs:fit:860:0:0:0/g:ce/aHR0cHM6Ly9jZG4u/ZGVlcHNlZWsuY29t/L2FwaS1kb2NzL2Rl/ZXBzZWVrX3IxX2Jl/bmNobWFyay5wbmc align="left")

**Getting Started:**

1. **Install Ollama:**
    
    * **macOS:** Open your terminal and run: `brew install ollama`
        
    * **Linux:** Open your terminal and run: `curl -fsSL https://ollama.ai/install.sh | sh`
        
    * **Windows:** Follow the installation instructions from the official Ollama website.
        
2. **Pull DeepSeek:**
    
    * Open your terminal and run: `ollama pull deepseek-r1:1.5b`
        
3. **Run DeepSeek:**
    
    * In your terminal, run: `ollama run deepseek-r1:1.5b`
        
    
    for example:
    
    ![](https://cdn.hashnode.com/res/hashnode/image/upload/v1740462281698/c6728eda-1c27-44cb-8c1e-3911234171cc.png align="center")
    
4. **Interact in the Terminal:**
    
    * Ask questions or provide instructions. Example: "Write a python function that reverses a string."
        
5. **Stop DeepSeek:**
    
    * Press `Ctrl+D` (or `Cmd+D` on macOS).
        
6. **Using DeepSeek in a Script:**
    
    * You can integrate DeepSeek into your Python scripts: Python
        
        ```plaintext
        import ollama
        response = ollama.chat("deepseek-r1:1.5b", "Hello, how can you assist me?")
        print(response)
        ```
        
7. **Using a Web UI:**
    
    * For a better user experience, consider using Open WebUI: [https://github.com/open-webui/open-webui](https://github.com/open-webui/open-webui)
        
    * After installation, you can access the UI at `http://localhost:11434/api/chat`.
        

**Important Considerations:**

* Running LLMs locally requires sufficient hardware resources.
    
* Experimenting with different prompts will help you achieve better results.
    

DeepSeek and Ollama together provide a powerful and accessible way to use LLMs locally.

Models like deepseek will keep coming in the market, it’s us (coders) who have to improve :D, do share your opinion on this statement in the comments!!  
  
Till then, explore, take care, learn and enjoy 🔥

### Extra links:

**Discord:** [https://discord.](https://discord.com/invite/hackunited)gg/hackunited

**Thanks a lot for checking this out, we hope that you have fun! :D**