What in tarnation is going on with the cost of compute

r/LocalLLaMA
AI Hardware

Does anyone know? I can’t even find a server gpu