In this presentation I give an overview of the hardware you can buy on the market today for running large language models locally. Specifically I discuss the following systems: * Apple Mac Studio M3 Ultra 512 GB & Apple MacBook Pro M4 Max * NVIDIA DGX Spark and DGX Station (also known as ASUS Ascent GX10) * NVIDIA RTX 6000 Blackwell Generation, RTX 5090, RTX 4090, RTX 3090 * NVIDIA Jetson AGX Orin * Intel project Battlematrix * AMD Ryzen AI Max+ Pro 395 (Strix Halo) * AMD Radeon Pro W7900 All of these systems are presented with their purchase prices, some general insights, and also I predict token/s values for two different Llama models (Llama 3.3 70B Q8 and LLama 3.1 405 B 4bit quant). Will it fit? I encourage our viewers to comment below about any hardware I might have missed out on, or if your token/s values / purchase price experience varies. Subscribe to our newsletter to get this presentation: And if you want a deep dive with your company, book a free 30 min strategy call now (limited seats available each month): Subscribe to my channel for more tips on AI for managers, entrepreneurs and business people, upcoming AI tools which will save you time and make you money.