<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>Forem: Raja</title>
    <description>The latest articles on Forem by Raja (@marimuthuraja).</description>
    <link>https://forem.com/marimuthuraja</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F1065502%2Ffec85f51-729e-4cee-adb1-886e123d9cb1.jpg</url>
      <title>Forem: Raja</title>
      <link>https://forem.com/marimuthuraja</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://forem.com/feed/marimuthuraja"/>
    <language>en</language>
    <item>
      <title>Install DeepSeek in Your Local Computer</title>
      <dc:creator>Raja</dc:creator>
      <pubDate>Sun, 02 Feb 2025 11:39:00 +0000</pubDate>
      <link>https://forem.com/marimuthuraja/install-deepseek-in-your-local-computer-42j1</link>
      <guid>https://forem.com/marimuthuraja/install-deepseek-in-your-local-computer-42j1</guid>
      <description>&lt;p&gt;DeepSeek, DeepSeek, DeepSeek... We can here this word from small youtube channels to American President in the recent times. It is open source, I thought Ok We can try this. I see the model have 1.5B version which is not that much efficient but, we can give a try on that ( I don't have GPUs to run the 671B 😅). Let's install 1.5b version in my desktop.&lt;/p&gt;

&lt;p&gt;Before the installation, i want to introduce one thing called, &lt;a href="https://ollama.com/" rel="noopener noreferrer"&gt;OLLAMA&lt;/a&gt; which is a software to run any open source large language models in our local system or even in a server. There are some alternatives doing the same job but ollama is very easy to setup and it is too easy to work with.&lt;/p&gt;

&lt;h2&gt;
  
  
  Setup Ollama
&lt;/h2&gt;

&lt;p&gt;This is an easy process, just go to &lt;a href="https://ollama.com/" rel="noopener noreferrer"&gt;https://ollama.com/&lt;/a&gt; and donwload it, it is available for Mac, linux and windows, go ahead, download and install it.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fy9v848nyzbdw57k7o2j7.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fy9v848nyzbdw57k7o2j7.png" alt="Installing Ollama" width="800" height="590"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Once the installation is done, verify it is installed properly by running the command in your terminal&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;ollama --version
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fmyrgyilf9u07vtb9h428.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fmyrgyilf9u07vtb9h428.png" alt="Ollama Version" width="371" height="89"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;As you can see, our Ollama installed successfully.&lt;/p&gt;

&lt;h2&gt;
  
  
  Installing DeepSeek:
&lt;/h2&gt;

&lt;p&gt;Deepseek has various model versions from 1.5B to 671B , you can see the list of available models &lt;a href="https://ollama.com/library/deepseek-r1" rel="noopener noreferrer"&gt;here&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;To install DeepSeek in our local machine, open your terminal and run the following command&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;ollama run deepseek-r1:1.5b
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;This will pull the deepseek model file and run the model&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F0f5d2soa4hz49kzztehn.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F0f5d2soa4hz49kzztehn.png" alt="Ruuning DeepSeek in Local Machine" width="516" height="133"&gt;&lt;/a&gt;&lt;br&gt;
that's it now it is installed in our local system 🥳🎉&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F36wah03c1j0nt5p5dn8j.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F36wah03c1j0nt5p5dn8j.png" alt="Deep Seek Response" width="442" height="186"&gt;&lt;/a&gt;&lt;/p&gt;
&lt;h2&gt;
  
  
  Bonus Tip
&lt;/h2&gt;

&lt;p&gt;Everything looks good for now. But it's kind of boring using the terminal. It will look cool if we have a chatgpt kind of ui right?, I got you. There is an open source repository from &lt;a href="https://github.com/open-webui/open-webui" rel="noopener noreferrer"&gt;OpenWebUI&lt;/a&gt; which fulfills our need, so why can't we use that? let's install it. &lt;/p&gt;

&lt;p&gt;If you have docker installed, it is a straight forward way.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;docker pull ghcr.io/open-webui/open-webui:main
docker run -d -p 3000:8080 -v open-webui:/app/backend/data --name open-webui ghcr.io/open-webui/open-webui:main
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;If you don't have docker installed, don't worry they published this as a pip package, but make sure you have python 3.11 installed. you can run the following command to install webui&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;pip install open-webui
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Once the installation is done, you can run the webapp using the following command&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;open-webui serve
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F4wj2fc618yyi6xuxcvvw.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F4wj2fc618yyi6xuxcvvw.png" alt="Open Web UI Installation" width="800" height="461"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Now our Chatgpt like webapp is ready&lt;br&gt;
open &lt;a href="http://localhost:8080" rel="noopener noreferrer"&gt;http://localhost:8080&lt;/a&gt; it will prompt for signup. Once signup is done you can see the Chatgpt like UI&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Foo8i7feu88jcok2lwdlr.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Foo8i7feu88jcok2lwdlr.png" alt="Open Web UI Interface" width="800" height="452"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Now we can play with it🥳🥳&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F3hpq3uak0avsrmmtt6sw.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F3hpq3uak0avsrmmtt6sw.gif" alt="deep seek openweb ui celebration" width="498" height="278"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;If you like my blogs please do follow me on &lt;a href="//www.linkedin.com/in/marimuthu-raja"&gt;Linkedin&lt;/a&gt; and &lt;a href="https://www.instagram.com/_mmraja_/" rel="noopener noreferrer"&gt;Instagram&lt;/a&gt;&lt;/p&gt;

</description>
      <category>ai</category>
      <category>deepseek</category>
      <category>python</category>
      <category>machinelearning</category>
    </item>
  </channel>
</rss>
