What's the accuracy tradeoff for the INT4 model vs non-quantized Llama 3.1 405b?
· Sign up or log in to comment