Singapore-founded AI cloud provider Aolani plans an AI factory in the Philippines for early 2027 as part of a new compute collaboration with NVIDIA. The company expects to deploy 22,000 GPUs across facilities in the Philippines and Malaysia.
Aolani announced the collaboration on September 21. It said the expansion would bring its total deployment to more than 48,000 GPUs and its total AI factory capacity above 100MW. The company also plans to deploy NVIDIA Blackwell Ultra GPUs.
The announcement does not specify that all 22,000 GPUs will be Blackwell Ultra models. Aolani also did not say how the GPUs will be divided between the Philippines and Malaysia, or disclose the Philippine facility’s location, individual power capacity, or project cost.
Aolani described the Philippine deployment as being “among the first large-scale AI infrastructure platforms” in the country.
The Philippines already has infrastructure designed for AI workloads. VITRO Sta. Rosa in Laguna lists 36MW of IT load capacity and describes itself as the country’s first AI-ready hyperscale data center. ePLDT also runs its GPU-as-a-Service offering on NVIDIA GPU infrastructure hosted at VITRO Sta. Rosa.
Despite the name, an AI factory is not a manufacturing plant. NVIDIA uses the term for specialized computing infrastructure optimized for AI workloads, covering processes such as data ingestion, model training, fine-tuning, and high-volume inference.
NVIDIA’s current partner directory lists Aolani Holdings Limited as a Cloud Partner, with competencies in compute and NVIDIA Enterprise Software. Aolani said it has been an NVIDIA Cloud Partner since 2023.
Under the collaboration, Aolani said it will provide NVIDIA-powered cloud services to AI companies, enterprises, startups, developers, and institutions across Southeast Asia. The agreement also includes a revenue-sharing and credit-support model intended to link infrastructure deployment with customer demand.