Install Ollama
Ollama is tool for running large language models locally. Download and install app for your operating system. Also you can download from GitHub.
Download models
After install in Models menu you can find commands for download models on you computer:
~5GB normal for 16GB operating memory. 14 billion parameters also ok but works slowly.
After install app you can use command ollama
in your cli:
You can pull and run models from list:
Show all models on your computer:
Example of usage
Start and try write prompt for example:
How does an aircraft engine work?
is a really hard question. LM DeepSeek is a thinking model and writes how it thinks before giving an answer.
Also you can see how memory using in Activity monitor at start and in the end of work.
Finally: