Here is how you can create your own offline AI and run it directly on your PC, or even from a USB drive.

To do this, you need an LLM file and a GGUF model file. Download the latest LLM file from the GitHub page (link provided below). You will also need the Qwen GGUF file, which you can download from the link below. Please note that the file size is about 2.5 GB.

Once both files are downloaded, keep them in the same folder.

Now, rename the LLM file and add the `.exe` extension to it. After that, open Notepad and type:

./llamafile.exe –server –model (its double dash for both server and model)

Then, copy and paste the exact model name after the `–model` parameter. Also, make sure the `.exe` file name is correct.

Now, save this file as a `.bat` file in the same folder.

To start the offline AI, simply double-click the BAT file we just created. It will provide you with a local link. Copy the link and paste it into your browser, and you will have your offline AI running directly from your PC.

You can ask questions, upload text files and PDFs, and experiment with different models and settings.

If you want to use another model, simply download the required GGUF file, update the model name in the BAT file, save it, and run it again.

Thank you very much!