4 ms·Run Llama locally on CPU with minimal API's in-between you and the model3 points by anordin95 2y ago