News
DealsCorvex agrees USD 103 million take-or-pay deal to perform AI inference with Mid-Atlantic GPUsSovereignNeysa scales AI infrastructure in India with AMD Instinct MI350X GPUs on TytanInferenceMicrosoft Foundry adds Baseten, DeepSeek AI inference (MSFT:NASDAQ)ClustersDataOne and Rowan University sign engineering co-op program agreement for DataOne's Vineland AI factoryClustersLook inside Cal Poly's $3 million AI Factory (Video)DealsOpenAI Taps AMD Chips for 457 MW of AI Capacity in Six-Year TensorWave DealClustersCal Poly launches an AI data center on campus. Take a look insideInferenceArista takes aim at the next AI network battleground: distributed inferenceDealsOpenAI Signs 457 MW AI Data Center Deal as AMD Expands Its Challenge to NvidiaClustersBehind Dell AI Factory鈥檚 NVIDIA & AMD Enterprise ExpansionDealsCorvex agrees USD 103 million take-or-pay deal to perform AI inference with Mid-Atlantic GPUsSovereignNeysa scales AI infrastructure in India with AMD Instinct MI350X GPUs on TytanInferenceMicrosoft Foundry adds Baseten, DeepSeek AI inference (MSFT:NASDAQ)ClustersDataOne and Rowan University sign engineering co-op program agreement for DataOne's Vineland AI factoryClustersLook inside Cal Poly's $3 million AI Factory (Video)DealsOpenAI Taps AMD Chips for 457 MW of AI Capacity in Six-Year TensorWave DealClustersCal Poly launches an AI data center on campus. Take a look insideInferenceArista takes aim at the next AI network battleground: distributed inferenceDealsOpenAI Signs 457 MW AI Data Center Deal as AMD Expands Its Challenge to NvidiaClustersBehind Dell AI Factory鈥檚 NVIDIA & AMD Enterprise Expansion
EveryNeocloud

Training clusters

Reserved clusters of thousands of GPUs on one fabric. 105 providers on EveryNeocloud so far.

A training cluster is a block of hundreds to hundreds of thousands of GPUs joined on one fast network, reserved by one customer for months or years. This is how frontier models are trained, and it is the core business of the largest neoclouds. Deals are signed as contracts worth billions of dollars, often before the data centre is built.

What makes a cluster

Inside each server, GPUs talk over NVLink; between servers they use InfiniBand or fast Ethernet in a layout called a fat tree, so any GPU can reach any other at full speed. Shared storage feeds training data, and software such as Slurm or Kubernetes schedules jobs and restarts them when a GPU fails, which in a cluster of 100,000 GPUs happens many times a day.

How it is sold

Customers sign take-or-pay contracts: they pay for the capacity whether or not they use it. Microsoft, OpenAI, Meta and Google have signed such deals with CoreWeave, Nebius, Nscale, IREN and others. The contracts let providers borrow against them to buy GPUs. Critics say this ties a small number of buyers and lenders together; providers say long contracts make the business stable.

Famous training clusters providers

Further reading

馃嚭馃嚫 United States and 1 moreCoreWeave US GPU cloud running AI data centres for hyperscalers and model labsNeocloud 路 Nvidia Hopper 路 2017 G42 馃嚘馃嚜 United Arab Emirates and 4 moreG42 Abu Dhabi AI holding group running cloud, data centres and national compute projectsNeocloud 路 Nvidia Blackwell 路 2018 馃嚥馃嚲 Malaysia and 4 moreYTL AI Cloud Malaysian AI cloud and supercomputer run from a solar-backed Johor data centre parkNeocloud 路 Nvidia Blackwell 路 2024 Megaspeed 馃嚫馃嚞 Singapore and 4 moreMegaspeed Singapore AI cloud firm investigated over Nvidia chip diversion to ChinaNeocloud 路 Nvidia Blackwell 路 2015 Firmus 馃嚘馃嚭 Australia and 2 moreFirmus Neocloud operator building immersion-cooled AI Factories across Asia-PacificNeocloud 路 Nvidia Blackwell Arc Compute 馃嚭馃嚫 United StatesArc Compute Turnkey NVIDIA GPU cluster design, deployment and management for AI workloadsNeocloud 路 Nvidia Blackwell 路 2018 GMI Cloud WorldGMI Cloud AI-native cloud for production inference, GPU clusters and agent runtimesNeocloud 路 Nvidia Hopper Lambda 馃嚭馃嚫 United StatesLambda AI-only GPU cloud running NVIDIA supercomputers for training and inferenceNeocloud 路 Nvidia Blackwell 路 2012 NHN Cloud 馃嚢馃嚪 South KoreaNHN Cloud South Korean cloud provider with large GPU clusters for AI trainingNeocloud 路 Nvidia Blackwell Solstice 馃嚭馃嚫 United StatesSolstice DOE AI supercomputer with 100,000 Nvidia Blackwell GPUs at ArgonneSupercomputer 路 Nvidia Blackwell WhiteFiber 馃嚭馃嚫 United StatesWhiteFiber AI data centres, private GPU clouds and reserved GPU capacity in North AmericaNeocloud 路 Nvidia Blackwell Microsoft Fairwater 馃嚭馃嚫 United StatesMicrosoft Fairwater Purpose-built AI supercomputer campus in WisconsinHyperscaler 路 Nvidia Blackwell 路 2025 Together AI 馃嚭馃嚫 United StatesTogether AI AI cloud running open models on demand, with GPU clusters and inference researchNeocloud 路 Nvidia Blackwell Crusoe 馃嚭馃嚫 United States and 3 moreCrusoe AI cloud and modular data centre operator that also builds its own power supplyNeocloud 路 Nvidia Blackwell Radiant 馃嚞馃嚙 United KingdomRadiant AI compute infrastructure company building repeatable 'AI factories' powered by its FlightDeck platform.Neocloud 路 Nvidia Blackwell Verda 馃嚝馃嚠 FinlandVerda European full-stack AI cloud offering GPU clusters, instances and model APIsNeocloud 路 Nvidia Blackwell RunPod 馃嚭馃嚫 United StatesRunPod AI developer cloud for training and inferenceNeocloud 路 Nvidia Blackwell Vast.ai 馃嚭馃嚫 United StatesVast.ai Marketplace for distributed GPU cloud computeMarketplace 路 Nvidia Blackwell 路 2016 Horizon 馃嚭馃嚫 United StatesHorizon Largest academic supercomputer in the United StatesSupercomputer 路 Nvidia Blackwell 路 2026 Stargate Norway 馃嚦馃嚧 NorwayStargate Norway 100,000 GPU AI infrastructure project in Northern NorwaySovereign cloud 路 Nvidia Blackwell 路 2025 Hon Hai Kaohsiung Super Comput 馃嚬馃嚰 TaiwanHon Hai Kaohsiung Super Computing Center Taiwan鈥檚 fastest AI supercomputer built by Foxconn and NvidiaSupercomputer 路 Nvidia Blackwell 路 2025 E2E Networks 馃嚠馃嚦 IndiaE2E Networks India's listed GPU cloud for AI training and inference.Neocloud 路 Nvidia Hopper 路 2009 Polarise 馃嚛馃嚜 GermanyPolarise German sovereign AI cloud running its own data centres and open-weight model APIsNeocloud 路 Nvidia Blackwell 路 2024

Photo credits on this page

The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.

The weekly digest

Every Monday: the provider of the day, new pages on EveryNeocloud, and the conferences and anniversaries of the week ahead.