Observed arrival · 2026-10-18
Rig Per Dollar: matching local models to hardware
A budget-and-model picker for comparing PCs and GPUs that can run local AI models.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- People choosing hardware for local LLMs
- Worth noticing
- Speed figures are labeled as memory-bandwidth estimates until measured results are available.
Field notes
The model list marks some entries as older, making the distinction between current and still-widely-run options visible. Hardware choices extend beyond discrete graphics cards to Apple Mac Studio configurations and a Framework Desktop listing with 192 GB of memory; the homepage also links to methodology and fully priced builds.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue