FitLLM is a free, accurate local-LLM fit calculator for NVIDIA and AMD GPUs and Apple Silicon Macs. Choose a model, weight quant, KV-cache quant and context length to see whether it fits your GPU's VRAM or your Mac's unified memory and how much is left. It correctly models the sliding-window and hybrid attention of Gemma 4 and Qwen 3.6 and the MLA compressed KV cache of GLM-5.2, GLM-4.7-Flash and the DeepSeek family — which naive VRAM calculators get wrong by multiples — and lets you paste any HuggingFace model. Every number is computed by an open-source MIT engine from official config.json values, with hardware specs cross-verified against two independent sources. 이 LLM, 내 GPU·맥에 들어갈까?