What is an AI datacenter?

An AI datacenter is a facility purpose-built or retrofitted for GPU and accelerator workloads — typically at rack densities above 30 kW, with liquid-cooling readiness and high-bandwidth fabric (InfiniBand NDR, NVLink, or Spectrum-X). At least 40% of IT load runs AI training or inference on GPUs such as NVIDIA H100, H200, B200, or GB200 NVL72.

Which datacenter has the most GPUs?

As of 2026, xAI Colossus in Memphis, Tennessee holds the title of the world's largest single-site GPU deployment, with more than 200,000 NVIDIA H100 and H200 GPUs in one InfiniBand domain. Meta Hyperion (Louisiana) and OpenAI Stargate Abilene (Texas) are scaling toward gigawatt-class deployments that will overtake it within the next 18 months.

What is the largest AI cluster?

xAI Colossus is the largest contiguous AI training cluster in 2026, exceeding 200,000 NVIDIA H100/H200 GPUs on a single InfiniBand NDR 400G fabric. Meta's Hyperion build and OpenAI's Stargate program are constructing multi-site fabrics that will exceed one million accelerators across coupled campuses by 2027.

Which GPU cloud is cheapest for H100?

On-demand: Together AI ($1.75/hr) and Voltage Park ($1.89/hr) lead. Reserved 1-year: Voltage Park ($1.20/hr) and Crusoe ($1.39/hr) are typically lowest. AWS p5 sits at the top of the price stack at ~$4.10/hr on-demand.

How is Datacenter.computer different from a datacenter directory?

Datacenter.computer is an agent-readable intelligence layer, not a directory. It normalizes pricing, capacity, power, fabric, and latency across 240+ facilities and 40+ operators into a single queryable graph for AI agents, traders, and procurement teams.

How many AI datacenters are in USA?

78 indexed AI-capable datacenters operate in USA, totaling 21,051 MW of contracted IT capacity.

Which operators run datacenters in USA?

AWS, Aligned, Applied Digital, Cologix, CoreSite, CoreWeave, Crusoe, CyrusOne, Dell, Digital Realty, Equinix, Flexential, Foundry, G42 Cloud, GlobalFoundries, Google Cloud, HPE, IREN, Intel, Lambda, Meta, Micron, Microsoft Azure, NVIDIA/Foxconn, NVIDIA/Wistron, OpenAI/Oracle, Oracle Cloud, QTS, Rakuten, SK hynix, SMC, Stack, Supermicro, TSMC, TensorWave, Vantage, Voltage Park, xAI.

Which cities in USA have AI datacenters?

Abilene, TX, Allen, TX, Ashburn, VA, Aurora, CO, Boardman, OR, Boise, ID, Boydton, VA, Carrollton, TX, Champaign, IL, Chandler, AZ, Charlotte, NC, Cheyenne, WY, Chicago, IL, Childress, TX, Clay, NY, Council Bluffs, IA, Dallas, TX, Douglasville, GA, Eagle Mountain, UT, Elk Grove, IL, Ellendale, ND, Glendive, MT, Goodyear, AZ, Henderson, NV, Hillsboro, OR, Houston, TX, Kansas City, MO, Las Vegas, NV, Los Angeles, CA, Malta, NY, Manassas, VA, Memphis, TN, Mesa, AZ, Mt Pleasant, WI, Nashville, TN, New Albany, OH, Phoenix, AZ, Plano, TX, Pryor, OK, Quincy, WA, Richland Parish, LA, San Antonio, TX, San Jose, CA, Santa Clara, CA, Sterling, VA, Suwanee, GA, Temple, TX, The Dalles, OR, Tucson, AZ, Weehawken, NJ, West Lafayette, IN, Williston, ND.

[DC.COMPUTER] Layer 1 · Country · 38 operators

AI Datacenters in

Every indexed AI and GPU datacenter operating in USA — hyperscaler regions, neoclouds, sovereign clusters, and colocation campuses.

USA

[AI Summary]

USA hosts 78 AI-capable datacenters totaling 21,051 MW of contracted IT capacity. The market spans 38 operators (AWS, Aligned, Applied Digital, Cologix, and more) across 52 cities, covering hyperscaler regions, GPU-native neoclouds, sovereign AI platforms, and colocation campuses.

Facilities

21,051 MW

Total IT

Operators

Cities

[Facilities · 78]

Facility	Operator	Location	GPU	MW	PUE
CoreWeave NJ01	CoreWeave	Weehawken, NJ, USA	H100/H200	28	1.18
CoreWeave VA02	CoreWeave	Ashburn, VA, USA	H200/B200	60	1.15
CoreWeave IL01	CoreWeave	Chicago, IL, USA	H100	32	1.22
CoreWeave CA01	CoreWeave	Las Vegas, NV, USA	H100	18	1.28
Lambda SF1	Lambda	San Jose, CA, USA	H100	12	1.24
Lambda TX1	Lambda	Allen, TX, USA	H200	24	1.21
Crusoe MT-01	Crusoe	Glendive, MT, USA	H100	40	1.16
Crusoe Abilene	Crusoe	Abilene, TX, USA	B200	200	1.14
TensorWave AZ1	TensorWave	Phoenix, AZ, USA	MI300X	15	1.25
Voltage Park TX	Voltage Park	Dallas, TX, USA	H100	22	1.23
AWS us-east-1	AWS	Ashburn, VA, USA	H100/Trainium2	800	1.15
AWS us-west-2	AWS	Boardman, OR, USA	H100	600	1.13
Azure East US	Microsoft Azure	Boydton, VA, USA	H100/B200	700	1.17
Azure West US 3	Microsoft Azure	Goodyear, AZ, USA	H100/MI300X	450	1.19
GCP us-central1	Google Cloud	Council Bluffs, IA, USA	TPU v5/H100	500	1.10
GCP us-west4	Google Cloud	Henderson, NV, USA	TPU v5p	250	1.12
OCI Phoenix	Oracle Cloud	Phoenix, AZ, USA	H100/B200	120	1.20
Meta Temple	Meta	Temple, TX, USA	H100 (Grand Teton)	400	1.09
xAI Colossus	xAI	Memphis, TN, USA	H100/H200 · 200K	250	1.18
Equinix DC11	Equinix	Ashburn, VA, USA	Mixed	35	1.40
Digital Realty IAD	Digital Realty	Ashburn, VA, USA	Mixed	100	1.38
AWS us-east-2	AWS	New Albany, OH, USA	H100/Trainium2	550	1.14
AWS us-west-1	AWS	San Jose, CA, USA	H100	220	1.20
AWS Trainium Cluster	AWS	Aurora, CO, USA	Trainium2	180	1.18
Azure South Central US	Microsoft Azure	San Antonio, TX, USA	H100/B200/GB200	500	1.18
Azure West US 2	Microsoft Azure	Quincy, WA, USA	H100/B200	600	1.12
Azure North Central US	Microsoft Azure	Chicago, IL, USA	H100	300	1.20
Azure Stargate Mt Pleasant	Microsoft Azure	Mt Pleasant, WI, USA	GB200 NVL72	1000	1.10
GCP us-west1	Google Cloud	The Dalles, OR, USA	TPU v5p/H100	400	1.09
GCP us-east4	Google Cloud	Ashburn, VA, USA	TPU v5/H100	350	1.11
GCP Stargate OKC	Google Cloud	Pryor, OK, USA	TPU v6 Trillium	600	1.10
OCI Ashburn	Oracle Cloud	Ashburn, VA, USA	H100/H200	150	1.18
OCI Chicago	Oracle Cloud	Elk Grove, IL, USA	H100	90	1.22
Meta Eagle Mountain	Meta	Eagle Mountain, UT, USA	H100 (Grand Teton)	360	1.08
Meta Richland Parish	Meta	Richland Parish, LA, USA	GB200 (planned)	2000	1.08
Meta Mesa	Meta	Mesa, AZ, USA	H100	250	1.10
xAI Colossus 2	xAI	Memphis, TN, USA	GB200 · 1M planned	1200	1.18
OpenAI Stargate Abilene	OpenAI/Oracle	Abilene, TX, USA	GB200 · 64K+	1200	1.13
CoreWeave Plano	CoreWeave	Plano, TX, USA	H200	30	1.19
CoreWeave Atlanta	CoreWeave	Douglasville, GA, USA	H200/B200	100	1.16
CoreWeave Hillsboro	CoreWeave	Hillsboro, OR, USA	H100	45	1.17
Lambda Phoenix	Lambda	Phoenix, AZ, USA	H200	18	1.23
Lambda Kansas City	Lambda	Kansas City, MO, USA	B200	30	1.20
Crusoe Williston	Crusoe	Williston, ND, USA	H100	35	1.15
Crusoe Wyoming	Crusoe	Cheyenne, WY, USA	B200	50	1.14
Applied Digital Ellendale	Applied Digital	Ellendale, ND, USA	H100/H200	400	1.15
IREN Childress	IREN	Childress, TX, USA	B200 (planned)	600	1.16
TensorWave Tucson	TensorWave	Tucson, AZ, USA	MI325X	24	1.24
Foundry Texas	Foundry	Houston, TX, USA	H100	18	1.25
Rakuten Cloud NC	Rakuten	Charlotte, NC, USA	H100	20	1.30
Sustainable Metal Cloud	SMC	Champaign, IL, USA	H100 (immersion)	14	1.05
Equinix DC15	Equinix	Ashburn, VA, USA	Mixed	42	1.38
Equinix SV5	Equinix	Santa Clara, CA, USA	Mixed	28	1.36
Equinix CH2	Equinix	Chicago, IL, USA	Mixed	30	1.40
Digital Realty CLT	Digital Realty	Charlotte, NC, USA	Mixed	60	1.36
Digital Realty DFW	Digital Realty	Dallas, TX, USA	Mixed	85	1.34
QTS Manassas	QTS	Manassas, VA, USA	Mixed	180	1.35
QTS Atlanta-Suwanee	QTS	Suwanee, GA, USA	Mixed	110	1.38
Cyxtera DC2 (Cologix)	Cologix	Sterling, VA, USA	Mixed	40	1.42
Aligned IAD-04	Aligned	Sterling, VA, USA	Mixed	96	1.28
CyrusOne Chandler	CyrusOne	Chandler, AZ, USA	Mixed	120	1.34
Stack Infrastructure DFW	Stack	Carrollton, TX, USA	Mixed	100	1.33
Vantage VA1	Vantage	Ashburn, VA, USA	Mixed	192	1.30
Flexential Hillsboro	Flexential	Hillsboro, OR, USA	Mixed	36	1.32
CoreSite LA1	CoreSite	Los Angeles, CA, USA	Mixed	22	1.45
G42 Condor Galaxy 1 (USA)	G42 Cloud	Santa Clara, CA, USA	Cerebras CS-3	120	1.18
TSMC Arizona Fab 21	TSMC	Phoenix, AZ, USA	N4/N3 (AI accelerators)	500	1.35
SK hynix Indiana	SK hynix	West Lafayette, IN, USA	HBM packaging	300	1.34
Micron Boise HQ Fab	Micron	Boise, ID, USA	HBM3E	250	1.36
Micron Clay (planned)	Micron	Clay, NY, USA	DRAM/HBM (planned)	1000	1.35
Intel Ocotillo Fab 52	Intel	Chandler, AZ, USA	Intel 18A (Gaudi/Panther)	600	1.34
Intel Ohio One	Intel	New Albany, OH, USA	Intel 14A (planned)	800	1.32
GlobalFoundries Fab 8	GlobalFoundries	Malta, NY, USA	12nm/22FDX (edge AI)	200	1.38
Supermicro San Jose	Supermicro	San Jose, CA, USA	AI server assembly	30	1.40
Dell AI Factory TN	Dell	Nashville, TN, USA	PowerEdge XE9680 assembly	22	1.40
HPE Houston AI Server	HPE	Houston, TX, USA	Cray EX assembly	18	1.42
NVIDIA AI Supercomputer Houston	NVIDIA/Foxconn	Houston, TX, USA	Blackwell systems	50	1.40
NVIDIA Blackwell Packaging Dallas	NVIDIA/Wistron	Dallas, TX, USA	GB300 packaging	30	1.42

[Knowledge Graph]

Company → Datacenter → GPU → Customer → Industry

Every entity on this site is cross-linked. Follow the graph from operators down to specific facilities, GPU clusters, customers, and sectoral context.