Guides
Running AI on your own machine
Plain-language guides on VRAM, quantization, Mac versus PC, and which models are worth your disk space.
Basics · 5 min
How much VRAM do you need to run an AI model locally?
→The three things that take memory, how much of your VRAM is really usable, and a quick table from 4 GB to 48 GB.
Basics · 4 minQ4, Q5, Q8, FP16: which quantization should you download?
→What the file names mean, why bigger-at-Q4 beats smaller-at-Q8, and the default most people should pick.
Hardware · 6 minMac or Nvidia for local AI? The honest comparison
→Capacity versus speed, with real bandwidth numbers, and a table of which to buy for which goal.
Picks · 5 minThe best local AI models for an 8 GB graphics card
→What to install on the most common gaming GPU, by job: chat, code, reasoning and images.