llmrequirements.com maps local LLMs to the hardware that can actually run them

Maker truetotosse built a picker that lets you go in either direction: choose a model and see what hardware you need, or enter your hardware and see which LLMs will run on it. For indie developers trying to self-host inference without burning money on cloud APIs, having a quick hardware-to-model compatibility reference removes a lot of trial-and-error.