Democratizing LLMs - Running LLMs in Google Colab and CPU

My talk is on democratizing large language models by running them live with resources like Google Colab and system CPU., which are usually not used to run LLMs. HuggingFace and the open-source community have made it very easy for ordinary people to connect to state-of-the-art LLMs using quantization.

My plan is to show how easy it is to run a LLM in Colab with a live working demo. While also diving deep into the theory of these quantization methods. But really, all of this is to showcase the power of open-source and what happens when a diverse community of people come together to build something awesome.