Pretrain, finetune and serve LLMs on Intel platforms with Ray - View it on GitHub