Akamai publicly discloses technical specifications and details of its AI cluster infrastructure service offering.

Akamai Technologies, Inc. (NASDAQ: AKAM) has disclosed additional details on its recent four-year, $200 million service agreement with a major U.S. tech company at the forefront of the AI revolution for high-performance AI compute. Under the terms of the deal, the customer will utilize a multi-thousand NVIDIA Blackwell GPU cluster hosted in a data center designed for efficient, high-density power capacity, along with other cloud infrastructure services on Akamai's distributed cloud platform.

The GPU cluster represents one of the world's largest NVIDIA Blackwell RTX PRO 6000 Server Edition clusters at scale. It is powered by an AI-optimized Ethernet networking platform that enables non-blocking, lossless, and high-performance connectivity for large-scale AI factories and GPU-accelerated computing. The infrastructure also leverages a high-performance, parallel file storage platform featuring NVMe-over-Fabric to support linear scalability in AI and HPC workloads.

"This $200 million commitment is a powerful validation of our differentiated strategy to build a definitive global platform for the entire AI lifecycle," said Akamai COO and GM, Cloud Technology Group, Adam Karon. "We are providing an essential foundation for their AI workloads — a dedicated multi-thousand NVIDIA Blackwell GPU cluster integrated with the world's most distributed cloud — to deliver the predictable, reliable performance their end-users expect."

The deal underscores enterprise demand for Akamai's integrated AI development and deployment platform. It follows several strategic moves by Akamai to expand its AI capabilities, including a rapid expansion of its global IaaS footprint to 41 datacenters leveraging relationships with many data center partners, and expansion of its Managed Container Service. In October 2025, Akamai announced Akamai Inference Cloud, bringing AI inference closer to users and devices. Most recently, the company announced the acquisition of thousands of NVIDIA Blackwell GPUs to bolster its global distributed cloud infrastructure and create a unified platform for AI R&D, fine-tuning, and post-training optimization.

Akamai disclosed information about this service agreement during its Q4 and FY25 conference call on February 19, 2026.