llms-locally

1 месяц назад

This video explores the feasibility of running large language models (LLMs) on various hardware, ranging from a $50 Raspberry Pi to a $50,000 AI workstation. The presenter tests different models and configurations, showcasing the performance and limitations of each setup. The video aims to provide insights into the hardware requirements for running LLMs locally.

3 месяца назад

This video provides a guide to the best consumer GPUs for running large language models locally. The video focuses on Nvidia GPUs due to their superior support for AI software. The presenter recommends prioritizing GPUs with the most VRAM, as this is crucial for running large models smoothly.