1. Access https://ollama.com/ Official website

2. ClickDownloadCorresponding to myselfComputerSystem: My computer runs Windows, so I downloaded the Windows version of the Ollama program

3. DownloadInstallationOllama

After downloading the installation package, click install to install

By default, C drive installation is set to ensure there is free space on the C drive

4. Open the cmd command prompt

Enter the command and wait for the download and installation to complete

ollama run deepseek-r1:14b

This is the 14b command. Modify the model with different parameters and modify the numbers in the command to match them

At this point, the large model has been successfully installed and deployed

5. For greater convenience, we need a Web UI interface and moreTutorialI'll teach you how to use Docker. For those without Docker, it's inevitable to end up making dumplings just for a bunch of vinegar. This tutorial recommends installing Page Assist – local AI The model's Web UI browserPlugin

If you have magical conditions, you can visit directly:Page Assist – Web UI for native AI modelsOne-click installation access

If you don't have magical conditions, you can:https://pan.quark.cn/s/bbfe056725fdObtain the offline installation package

All resource packages used in this tutorial

Mainly to make it convenient for people without internet access to download it offline

https://pan.quark.cn/s/e23c61c2b371