AI Compute RadarAll GPU classes · GPU prices per hour
Hyperstack GPU prices per hour
What Hyperstack lists per GPU-hour right now for the classes it sells, each with the instance the price sits in, and what the list price did over the last weeks. Read four times a day; described, never forecast.
Radar Rent Index
What Hyperstack prices did
The same rule applied to Hyperstack's own daily series: the current window against the one before it, and against the first week collected. A class with too few days says so instead of showing a change.
| GPU | This week | vs. last week | Since start | Offers | Day band $/h |
|---|---|---|---|---|---|
| A100latest day Nov 2 | $1.35median of 7 days | —collecting | —collecting | —no comparison yet | —min–p75 of the latest day |
| H100latest day Nov 2 | $2.50median of 7 days | —collecting | —collecting | —no comparison yet | —min–p75 of the latest day |
| H200latest day Nov 2 | $3.99median of 7 days | —collecting | —collecting | —no comparison yet | —min–p75 of the latest day |
| B200latest day Nov 2 | $6.00median of 7 days | —collecting | —collecting | —no comparison yet | —min–p75 of the latest day |
No day was excluded. A 30-day change appears once thirty days are on record, from Nov 25.
A day is the median of its hourly medians, counting only hours with at least five offers (verified hosts on Vast.ai, free hosts on Clore.ai). A week is the median of its days and needs at least four of them. Words follow fixed thresholds: up or down from ±8 %, sharply from ±20 %; offers up or down from ±15 %. How the index is made →
Price history
Reading Hyperstack's prices
As of today's readingsOn November 2, 2026 Hyperstack listed 4 of the 8 tracked classes: A100 $1.35, H100 $2.50, H200 $3.99, B200 $6.00 per GPU-hour. It was the cheapest competing source on 1 of them (H100). Its widest distance to the cheapest reading was on the A100: $1.35 against $0.73 at Vast.ai, 1.8 times as much. Hyperstack has been read here since October 27, 2026.
Hyperstack is a GPU cloud run by NexGen Cloud that prices per card and bills per minute: each GPU comes with a ceiling of CPU cores and memory, and an instance is as many cards as you take. Its public pricing page needs no account. The number on this site is the on-demand price per GPU-hour from that page for each tracked class, which is the listed price itself, with nothing divided.
Four times a day the site reads the on-demand card of Hyperstack's pricing page and keeps the rows that belong to a tracked class: the H100 and A100 in their SXM, NVLink and PCIe rows, the H200 and the B200. The cheapest row of a class is its price and the row names the variant behind it. The plain L40 is a different card from the L40S and stays out, as do the reservation and spot cards further down the page.
Hyperstack's per-GPU figures tend to sit at the lower end of the cloud list prices on the board, closer to a single RunPod type than to an eight-card node, which is the product: one card with a fixed share of a host rather than a whole machine. Compare it with Nebius for the same per-card model, and with the marketplaces for what the same hour costs from a private host without support.
Price moves at Hyperstack, last 30 days
Any change of the list price between complete days; the big clouds move on the scale of months.
No move on record in the last 30 days.