Basic Desktop AI integration with Ollama
For software, info and contribution visit the github page.
https://github.com/wonka929/Ollama_AI_Desktop_Assistant
Small and capable LLM are emerging on the public domain.
I was very pleasentely surprised by the release of Phi3 from Microsoft which is a small size LLM that can be run on smartphones and has proven to be really good for its size.
You can find more info about Phi3 online.
I though it was time to start integrating a little bit more LLMs in my daily routine. I thought to how I usually use LLMs for and decided to create a completely LOCAL integration of Phi3 for my desktop.
Features:
- completely LOCAL (privacy and autonomy oriented)
- python libraries
- fast to use while doing your daily work
- scalable (you can use all the LLMs model supported by Ollama)
- adaptable (all the “functions” are nothing more than specific prompts)
What to do
Download Ollama
First of all you need to download and installa Ollama https://ollama.com/download
Follow instructions on the website to complete the installation.
Once installed, download an LLM model. I clearly chosed Phi3 and downloaded it running in terminal the command
ollama run phi3
Check that the model is working properly and continue.
Install Python3
Now you just have to download and install python3 if you haven’t already.
Download, install and include pip and tcl/tk in the installation.
Run the script
Now you should be able to just run the script by assigning him a shortcut on your OS.
Notes
Every OS has its own specific things to keep in mind for installation and configuration. Something more can be seen on the github page.
In any case these are just simple tasks everyone willing to integrate an LLM on its system should be able to perform.
