2 ms·And what hardware they needed to run the model, because that's the real pinch in local inference.by joquarky 9mo agoAnd what hardware they needed to run the model, because that's the real pinch in local inference.