LLM Memory (VRAM) Estimator

LLM Memory (VRAM) Estimator is a free browser tool that estimates how much GPU memory (VRAM) a local LLM needs for a given model size, quantization and context length. It shows the weights, KV cache and overhead separately and gives a fit verdict for common GPUs and Macs, so you can check whether a model will run on your hardware before you download it or buy a card. It runs entirely in the browser, needs no signup and is free to use (MIT licensed). Made by Yunus Emre Vurgun.