Running about raspberry

[nfaraji2002]

Hi,
Thanks for the great package.
Is it possible that we run any model of LocalAI on Raspberry PI 4?
Are the models in LocalAI able to run in a offline mode without internet?
How to customize a model on a specific language?

Thanks in advance,

[Aisuko]

Hi, @nfaraji2002. It is good for you to ask questions here. The first and second we can figure out after several months. There many people are interested in it. And for your third question, there is some useful repo you can use like 🤗transfomer, PEFT, ChatLLaMA, etc.

[mudler]

@nfaraji2002 rpi should work - however I didn't test it yet, there are container images too that you can try out.

Models are local - no remote connectivity is needed. For fine tuning a model: it really depends on the model family, there are various ways to do so, for instance: https://github.com/ZrrSkywalker/LLaMA-Adapter