Observed arrival · 2026-10-23
LLM Budget Lab: what will running a model cost?
A calculator compares the costs of running open-weight models on your own GPU or Mac, rented cloud hardware, or an API.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- People comparing local LLM hardware, cloud rental, and API costs
- Worth noticing
- The page flags cloud-rate figures as more than three weeks old and says provider pages should be checked before spending.
Field notes
The input flow includes quantisation and context length, with advanced fields for a card’s street price and monthly GPU load. Hardware choices span consumer GPUs, data-center accelerators, and Macs with up to 256GB unified memory. Its cloud-rate note says the underlying check is dated 2026-09-29, warns that provider pages are JavaScript-rendered, and says commissions are unverified.
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