Hyperscalers List GPUs at Nearly 2× the GPU-Cloud | Crypto News
The AI Compute Race report costs GPU compute exterior the greatest patrons: a three-year dedication cuts the median price by 48%, while 17 of 22 tracked GPU fashions rose in price over the past yr
DALLAS, Texas – November 13, 2026 – Fluence has printed The AI Compute Race, a report that costs GPU cloud compute for every company that will not be a frontier lab or a hyperscaler. It was produced with research companions NEAR AI, io.internet, Boundless, Phala, Gensyn, Fortytwo, Theta, Compute Labs and GAIB. Its central discovering: where a company buys a GPU-hour units the price. Across H100, H200 and B200 fashions, hyperscalers record GPUs at 1.98 instances the GPU-cloud median, from 1.4 instances on the B200 to 2.8 instances on the H100. The same H200 lists from $4.08 to $10.60 an hour.
Commitment size issues virtually as a lot as supplier alternative. Against each supplier’s own on-demand price, a one-year reservation cuts the median price by 21% and a three-year time period by 48%. Spot capability lists at 53% of on-demand, for workloads that may be interrupted.
Buyers who count on GPU costs to fall might need to recheck their budgets. Seventeen of the 22 GPU fashions tracked by GetDeploying value more than a yr earlier, and the median H100 rents for 13% to 30% more, relying on the window measured.
The report also traces why costs are quickly rising:
- Demand moved to open fashions. Chinese fashions’ share of OpenRouter tokens among corporations in the US, Europe and the Global South rose from 6–13% to 57–67% in under eight months, according to CNBC. Open weights let any company with GPUs serve a near-frontier model on {hardware} it controls.
- The greatest patrons locked up the latest provide. The 4 largest hyperscalers information $720–740 billion of 2026 capital spending, and the 4 largest disclosed offers by AI labs and Meta with neoclouds complete about $233 billion, three of them “up to” quantities and a lot of it for capability not yet delivered.
- The GPU-hour is changing into a financial asset. CME, ICE and Nodal Exchange have announced GPU futures, each settling on a different price index, while lenders are already lending against GPUs: Lambda’s investment-grade-rated loan priced at 6.78% and Sharon AI’s, with no score disclosed, at 9.95%.
“Open-weight models are close behind the best closed ones, and companies are switching to them. Any company with GPUs can now serve a near-frontier model on hardware it controls. The GPUs are the hard part,” said Evgeny Ponomarev, Co-founder of Fluence.
The report closes with 4 methods to pay less for the same compute: examine the same chip across supplier sorts, commit when the workload is regular, match the GPU era to the model, and look past the hyperscalers, since a lot of the latest neocloud capability is contracted to labs years forward.
The full report is out there at fluence.ai/compute-race-2026.
About Fluence
Fluence Cloud is a GPU cloud supplier with capability from unbiased data facilities (fluence.ai). Founded in 2017 by Dmitry Kurinskiy, Evgeny Ponomarev and Tom Trowbridge, the company has raised $14 million from traders including 1KX, Multicoin, Distributed Global, Blockchange, Tiger Global and Protocol Labs.
Stay up to date with the latest trending crypto news! Visit our web site daily for the freshest Crypto news and content, fastidiously curated to keep you informed.
