Thank you so much. Thank you for showing up and coming with me to join this session. And what we plan to do today, over the next 20 minutes, is to really cover.there's a lot to cover about inference, but I would like to sort of tell you a little bit about how we approach open source inference from vLLM, which is the inference engine that lets you run any large language model on data center hardware. And really there to. hopefully this will be a start and a spark of discussion and inspiration for you all.And ...