Exascale | AI Compute Exchange The institutional exchange for AI Compute. GPU Data centers sell capacity, AI NeoLabs and NeoClouds buy it, and traders make the market, on one venue. Exascale is also an operator: it sources, installs and runs GPU capacity in U.S. colocation sites, and the cloud and model API above it. The venue • Exchange: spot, on-demand and 1 to 5 year reserved compute, priced against the Exascale GPU Index, its published index of GPU-hour prices • Supply: data centers onboard capacity with facility attestation, a read-only telemetry endpoint and per-second metered settlement The operator • GPU Cloud: NVIDIA H100, H200, B200, B300, GB200 and GB300, metered per second, with managed Kubernetes and Slurm • AI Models: OpenAI-compatible inference for open models across text, code, reasoning, vision, image, video, speech and embeddings • Data centers: sites in Dallas, North Texas and Houston, run end to end from sourcing, inspection, testing and burn-in through installation, networking, Kubernetes, Slurm and offtake • GPU Ownership: owners hold GPU servers without running them. Exascale sources the hardware and the site, installs and operates it, and contracts the capacity to offtakers on 1 to 5 year terms, for a share of the revenue A published index, terms from spot to five years, and attested, metered supply. Built by Hyperlink Technologies, Inc. exascale.ai
| Website | http://www.hyperlink.org |
| Employees | 15 (4 on RocketReach) |
| Founded | 2022 |
| Industry | Technology, Information and Internet |
Looking for a particular Hyperlink employee's phone or email?
Marta Parke is the Founder and CEO of Hyperlink.
4 people are employed at Hyperlink.
Hyperlink is based in San Francisco, California.