Run Llama locally on CPU with minimal API's in-between you and the modelgithub.com3 points·anordin95··0 commentsOpen articleSaveView on HN