KONST
AI Training CalculatorContact Us
EN
AIDC site | North America - United States

NVIDIA GB300 - United States

US-GB300-NVL72POC 10 to 12 weeks after signing

Location
United States
Availability
POC 10 to 12 weeks after signing
Cooling
DLC liquid cooling
Reference price
$5.30 USD / hr / GPU, starting price - Monthly
50NVL72 racks
3,600GPU
288GB HBM3e per GPUMemory per GPUabout 1,037TB total
InfiniBand XDR 800GGPU Fabric
Specifications

Full Specifications

GPU, host, network, and site conditions in one place.

COMPUTECompute

GPU model
NVIDIA GB300
GPU configuration
3,600 units4x GB300 per compute tray900 trays72 GPU per rack (NVL72)
Deployment scale
50 NVL72 racks3,600 GPU
Memory per GPU
288GB HBM3e per GPU
Total cluster memory
about 1,037TB total

HOSTHost

Processor
2x NVIDIA Grace (72-core Arm Neoverse V2) per tray
System memory
LPDDR5X SOCAMM
Storage
OS 2x 1.92TB M.2Data 4x 3.84TB E1.S

NETWORKNetwork

GPU interconnect
InfiniBand XDR 800G per GPUConnectX-8Quantum-X800 two-tier non-blocking fat tree
External network
1x BlueField-3 DPU (dual-port 400G) per tray

FACILITYSite and Delivery

Location
United States
Cooling
DLC liquid cooling
Availability
POC 10 to 12 weeks after signing

WORKLOADSuitable Workloads

Purpose
AI TrainingInference

Remaining specifications are provided when you request a quote. Source: GPU cluster specification sheet · 2026-11-16

Delivery schedule

From Signing to Go-Live

  1. 01Signing5 year term
  2. 02POCPOC 10 to 12 weeks after signing
  3. 03HandoverConfirmed with quote
  4. 04Go-liveConfirmed with quote

Prices, specifications, and availability are confirmed in the signed contract and do not reflect live inventory.

Location

United States

United States
Region
North America
Country / city
United States
Site code
US-GB300-NVL72
Sites in the same location
US-B300

Location is indicative, at country or city level.

Compare | Other sites

Sites With NVIDIA GB300

Compare scale, network, cooling, and delivery time side by side.

This is currently the only site with NVIDIA GB300. Other GPU models are listed below for comparison.

SiteScaleGPU memoryGPU FabricCoolingAvailabilityStarting price, USD / hr / GPU
US-GB300-NVL72United StatesThis page50 NVL72 racks
3,600 GPU
288GB per GPU
about 1,037TB total
InfiniBand XDR 800GDLC liquid coolingPOC 10 to 12 weeks after signing$5.30 from
TW-H100Taiwan New Taipei - NVIDIA H1003 nodes
24 GPU
80GB per GPU
1.92TB total
Optional 400GAir CoolingIn stock, immediate delivery
7 to 14 days after the 3rd unit is signed
$2.70 from
JP-Chiba-H100Japan Chiba - NVIDIA H1004 nodes
32 GPU
80GB per GPU
640GB per node
400G EthernetAir CoolingDelivery within 6 weeks of signing$2.21 from
CZ-H200-NVLCzechia - NVIDIA H20012 nodes
96 GPU
141GB per GPU
1,128GB per node
400G ConnectX-8Air CoolingLive in December 2026
Expandable to 24 nodes
$2.75 from
CZ-B300-200Czechia - NVIDIA B30025 nodes
200 GPU
288GB per GPU
about 57.6TB total
On requestTo be confirmed16 weeks after signing$4.80 from
IN-B300India - NVIDIA B3008 to 256 nodes
Configurable
288GB per GPU
2.3TB per node
InfiniBand NDR 400GAir Cooling8 to 10 weeks after signing$4.86 from
US-B300United States - NVIDIA B3008 to 96 nodes
First batch 512 GPU
288GB per GPU
about 147TB total in first batch
InfiniBand NDR 400GAir Cooling10 to 14 weeks after signing$4.96 from
FI-B300Finland - NVIDIA B300144 nodes
1,152 GPU
288GB per GPU
2.3TB per node
800G ConnectX-8DLC liquid coolingAbout 22 weeks after signing$5.30 from
KR-B300South Korea - NVIDIA B3008 nodes
64 GPU
288GB per GPU
about 18.4TB total
800G RoCEv2DLC liquid coolingLive in December 2026Contact for pricing
US-IOWA-B300United States Iowa - NVIDIA B30044 nodes
352 GPU
288GB per GPU
about 101.4TB total
800G ConnectX-8Air CoolingLive in mid-February 2027Contact for pricing
US-SEA-B300-32United States Seattle - NVIDIA B30032 nodes
256 GPU
288GB per GPU
about 73.7TB total
800G ConnectX-8Air CoolingLive in February 2027Contact for pricing
US-SEA-B300-64United States Seattle - NVIDIA B30064 nodes
512 GPU
288GB per GPU
about 147.5TB total
800G ConnectX-8Air CoolingLive in mid-March 2027Contact for pricing
FAQ | Common questions

Common Questions

How are the price and delivery schedule confirmed?
Prices and delivery schedules are for evaluation reference. Actual specifications, availability, and commercial terms are set by the signed contract. Prices, specifications, and availability are confirmed in the signed contract and do not reflect live inventory.
How does bare metal differ from GPU cloud?
Bare metal is a physical machine with no virtualization layer, so all performance is available, but you manage the operating system and cluster yourself. GPU cloud comes with the environment prepared and activates quickly, but resources are shared with other tenants. Training at long-term full load usually goes to bare metal.
Besides overseas GPU clouds, what should be considered when renting locally?
Three things: data residency, network latency, and service hours. When data must not leave the country, equipment location is a hard requirement. Local latency is lower than transoceanic links. The local team keeps the same working hours as the lessee, so fault resolution does not wait on time zones.
Can the GPU count be increased or reduced mid-term?
Adding GPUs usually means adding nodes, with pricing folded into the original contract and subject to capacity at the time. Removing GPUs is restricted by take-or-pay terms, so in practice the scale is usually adjusted at renewal.
Who owns the equipment, and what does the rental fee cover?
KONST keeps ownership of the equipment. The renter pays for the right to use it, which covers data center space, power, network, and operations. KONST bears the risks tied to purchase funding, residual equipment value, and generation changes.
From installation to ready for use, where does KONST's delivery scope end?
From the initial operating system installation, KONST sets up the machines, drivers, and base environment, and configures a separate network segment and a dedicated VPN. After that, the renter manages frameworks, scheduling, and data storage on its own.

Plan the Next Stage of Compute Together

Where Efficiency Leads.
Where AI Defines the Future.