> ## Documentation Index
> Fetch the complete documentation index at: https://rajanand.org/llms.txt
> Use this file to discover all available pages before exploring further.

# How to install DeepSeek in Windows and run locally?

<Steps>
  <Step title="Download and install ollama">
    Download using this [link](https://ollama.com/download)
  </Step>

  <Step title="Check whether docker is installed">
    If docker is installed, you will see the version number. if not, install [docker](https://www.docker.com/).

    ```bash theme={"system"}
    docker --version
    ```
  </Step>

  <Step title="Pull the latest image of Open WebUI">
    ```bash theme={"system"}
    docker pull ghcr.io/open-webui/open-webui:main
    ```
  </Step>

  <Step title="Create a volume for Open WebUI">
    Create a volume for Open WebUI using the following [command](https://docs.docker.com/engine/reference/commandline/volume_create/)

    ```bash theme={"system"}
    docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
    ```
  </Step>

  <Step title="Wait for a minute and the access Open WebUI">
    Go to `http://localhost:3000` in browser to access Open WebUI.You will need to create a admin account by providing email and password. Then you will see the chat interface. But there won't be any model.
  </Step>

  <Step title="Install DeepSeek model">
    There are different models available. You can install 8B Llama deepseek R1 by running the following command. Then, you can now use the chat interface to interact with the model. Although it was working, it was slow for me. It took about 6 minutes to get the response.

    ```bash theme={"system"}
    ollama run deepseek-r1:8b
    ```

    <img src="https://mintcdn.com/rajanand/rzSsh53dcoBK6ypQ/assets/ai/deepseek-localhost.png?fit=max&auto=format&n=rzSsh53dcoBK6ypQ&q=85&s=c281a52ba1698e926e4344c0ee82b087" alt="DeepSeek Chat interface" width="734" height="244" data-path="assets/ai/deepseek-localhost.png" />
  </Step>
</Steps>