Using local models

[chenxizhang]

This is thanks to Ollama, whose work has made it so easy for us to run models locally, even if you don't have a machine equipped with a GPU. Please visit https://ollama.ai to learn more, and Windows users can get a quick experience through https://ollama.com/blog/windows-preview. There are currently hundreds of open-source models available, and you can even build your own (by modifying existing models).

Here is an example.

chat -endpoint local -model llama2-chinese `
-system "You are a medical professor, please explain medical knowledge and concepts in a way that a primary school student can understand"