Observed arrival · 2026-10-12
Clock, for the 12-minute GPU gap
A marketplace that says it turns idle GPU capacity into inference requests, priced moment by moment and yielded back when the owner’s workloads need it.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- GPU fleet operators with idle capacity
- Worth noticing
- The page quotes a 10–15% fee on revenue generated and says observation starts with a read-only agent.
Field notes
The proposed workflow begins with a read-only agent that models the operator’s fleet before routing any requests. Operators are asked to choose a fleet-size range, from under 100 to more than 10,000 GPUs, and can set allowed models and price floors. Clock states that it takes 10–15% of revenue it generates; the page’s utilization comparisons are explicitly illustrative.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
◎NicheUnusually specific use
One card from the complete issue