I’m trying to setup simple LLM server on Ubuntu 22.04. My GPU is Nvidia GeForce GTX 1050 (2048MB memory). I’m using Cuda Toolkit 12.4. I downloaded pretrained microsoft-phi3-mini model from https://huggingface.co/microsoft/Phi-3-mini-128k-instruct. The download was successful, and authentication is not required (I am logged in anyways). When I run example, provided on huggingface: