As enterprise AI demand outpaces available supply, Aethir ACCELERATE expands into physical infrastructure deployment, bringing NVIDIA GPU capacity online faster.
Today we're announcing Aethir ACCELERATE, Aethir's program designed to bring GPU compute online in AI data centers across the United States and Europe. ACCELERATE has secured access to 10 sites totaling up to 20 megawatts of capacity. Building out those sites is projected to generate up to $700 million in contracts by the end of 2026, and over $2 billion in total contract value once all 10 sites are fully built out. This marks one of the largest moves into physical AI infrastructure ever made by a decentralized compute network.
ACCELERATE takes Aethir down the stack. Until now, Aethir has aggregated and orchestrated GPU compute from third-party providers across hundreds of thousands of GPU containers in more than 90 countries, one of the largest live GPU networks in the world. With ACCELERATE, Aethir will identify and develop opportunities to bring GPU compute capacity online.
Aethir ACCELERATE’s facilities are purpose-built for high-density compute and rapid deployment, engineered to support NVIDIA B300 and GB300 clusters ranging from 64 to 256 nodes, built for AI training and inference workloads, and designed to go from site selection to live capacity in months, not the multi-year timelines that define traditional data center construction.
Demand Is Shifting to Faster, More Flexible AI Infrastructure
The market is growing rapidly, and McKinsey projects AI data center power demand will expand from 44 gigawatts in 2025 to 156 gigawatts by 2030, a 3.5x expansion in just 5 years. But supply cannot keep up: up to half of the global capacity scheduled to come online in 2026 faces delays from permitting, grid connections, and local opposition. In this market, speed of delivery is the scarce asset.
The AI workload mix is shifting in the same direction. Deloitte expects inference to account for roughly two-thirds of all AI compute in 2026. Training rewards giant centralized campuses. Inference rewards something different: capacity that comes online fast, sits closer to customers, and scales in steps as usage grows.
This is the market ACCELERATE is designed to serve. By accessing midsized sites, Aethir ACCELERATE can build and deploy in months rather than years to meet the demands of today’s enterprise AI workloads.
Depending on the site and customer requirements, Aethir’s services will include securing GPU hardware, arranging access to data center capacity, and engaging with key infrastructure partners, including Axe Compute. Through its Axe Build business, Axe Compute designs, deploys, owns, and operates dedicated AI infrastructure. The company has built one of the sector’s fastest-growing neoclouds by signed contract value, publicly reporting more than $3 billion in signed contracts in 2026 alone and almost $400 million in customer prepayments received. This track record of converting committed demand into deployed infrastructure is precisely what Aethir’s ACCELERATE model requires.
Aethir With Existing Demand for Accelerate
Enterprise customers on Aethir's network are already seeking capacity that traditional providers cannot deliver at the speed they need. Every ACCELERATE site launches against that existing demand, and the team is actively working to service it. The project creates clusters to fulfill existing contracted demand.
"We've already secured the sites. Now we're filling them. Enterprise AI demand has outrun supply, and the industry's ability to build is the bottleneck," said Mark Rydon, Co-founder and Chief Strategy Officer at Aethir. "The winners won't be whoever plans the biggest campus. They'll be whoever delivers working capacity first. Companies don't want to wait years for compute. They want serious capacity, deployed fast, shaped to their workload, and our network tells us exactly what to build before we select a single site."
What Does This Mean for ATH Holders?
Alongside the inception of the ACCELERATE program, Aethir will be unveiling an updated policy towards IDC tokenomics. This policy will include a burn mechanism and a variable platform fee, allowing the protocol to be more resilient during times of price volatility, and enabling existing ATH holders to benefit as on-chain revenue grows.
What's Next?
The first wave of ACCELERATE contracts is expected to close in the coming months. We'll announce individual sites as agreements are finalized.
This is just the beginning.
Disclosure: Aethir Foundation is an Axe Compute shareholder. This article reflects Aethir's views and is not investment advice. For official company information, see Axe Compute's SEC filings (CIK 0001446159) and investors.axecompute.com.
Nothing in this article should be relied upon as a guarantee of future performance or results.
