Rackspace signs AMD GPU deal for 30 MW AI compute, contingent on financing and approvals

Rackspace Technology, Inc. (the "Company") issued a press release announcing that its subsidiary, Rackspace US, Inc., (together with the Company, "Rackspace") has signed a definitive GPU-as-a-Service Agreement - Master Terms and Conditions (the "GPUaaS Agreement") with Advanced Micro Devices, Inc. ("AMD") relating to the phased deployment of AMD Products (as defined below) in Rackspace's global data centers.

On June 16, 2026, Rackspace and AMD announced that they have entered into a definitive GPUaaS Agreement. The GPUaaS Agreement is intended to facilitate the phased deployment by Rackspace of AMD AI compute products (including AMD Instinct™ GPUs (e.g., MI355X, MI350P, and future successor chips) and AMD EPYC™ CPUs) (collectively, the "AMD Products"), across Rackspace's global data centers. Under the GPUaaS Agreement, Rackspace has agreed to dedicate, maintain and make available an aggregate of 30 MW of capacity within its data centers as an initial launch footprint exclusively for AMD Products, where AMD Products are fit for purpose and subject to certain financing, operational, and legal conditions (the "AMD Footprint Obligation").

The parties intend for the initial deployment to commence in late 2026 and for the balance of the contemplated 30 MW to be deployed in phases through 2028. There can be no assurances that Rackspace will undertake deployments in the anticipated amounts, with the anticipated capacity, within the anticipated timeframe, or at all. Deployments by Rackspace under the GPUaaS Agreement will require Rackspace to obtain additional financing. There can be no assurances that Rackspace will be able to obtain financing on commercially reasonable terms or in the amounts necessary to facilitate such deployments, if at all.

Each deployment under the GPUaaS Agreement is subject to agreement by the parties on the commercial terms applicable to such deployment, including pricing, term and financial parameters, and AMD has no obligation under the GPUaaS Agreement to agree to any particular deployment as being within the scope of the framework of the GPUaaS Agreement. If the parties agree that a deployment is within the scope of the GPUaaS Agreement, AMD has agreed to purchase residual unsold capacity from the applicable deployment, subject to the satisfaction of delivery and service availability requirements and an aggregate cap for each deployment. There can be no assurances that AMD will agree to any deployment by Rackspace being within the scope of the GPUaaS Agreement.

The GPUaaS Agreement remains in effect until the expiration or termination of all deployments thereunder. Either party may terminate an affected deployment upon the other party's uncured material breach or insolvency, if the go-live date for such deployment has not occurred by a specified outside date, or upon certain service level failures.

AMD (NASDAQ: AMD) and Rackspace Technology® (NASDAQ: RXT), a global enterprise AI infrastructure and solutions provider, today announced the signing of a definitive agreement for the phased deployment of an initial 30 MW footprint dedicated to AMD-based compute deployments across Rackspace's global data centers beginning in late 2026 through 2028.

At full deployment, 30 MW of dedicated AMD compute across Rackspace's footprint will represent meaningful capacity to serve regulated enterprise workloads, including healthcare providers who have expressed early interest in accelerated compute for clinical AI and inference at scale.

While the parties have executed a definitive agreement establishing a commercial framework for the collaboration, individual deployment authorizations are subject to separate execution and certain commercial terms, including pricing and financial parameters, remain subject to further agreement between the parties. Any third-party financing required to implement planned deployments is subject to availability on terms acceptable to Rackspace in its sole discretion.

This agreement will accelerate delivery of the four integrated capabilities announced with the MOU: Enterprise AI Cloud, Enterprise Inference Engine, Inference as a Service, and Bare Metal AMD Instinct, offering a complete, governed stack from bare metal compute through fully operated inference.