Indonesia’s AI infrastructure push now has billions of dollars in financing behind it. Zankore signed a senior term loan facility of up to $3.1 billion on Sept. 9 to support the deployment of Nvidia GPU infrastructure for its AI cloud platform.
The company is building an initial 100 MW of AI infrastructure and plans to reach approximately 200 MW in the first half of 2027. Zankore’s longer-term target is up to 1 GW of Nvidia AI computing capacity across its broader Southeast Asian platform. The rollout begins in Indonesia, but the company has not said that the full 1 GW will be built there.
The debt financing comes from Citi, ING, Natixis Corporate & Investment Banking, Qatar National Bank, and UOB, which are serving as senior mandated lead arrangers, underwriters, and bookrunners. Citi also served as Zankore’s exclusive debt adviser, according to the lending consortium’s legal adviser.
How Zankore plans to use the financing
Indosat Ooredoo Hutchison launched Zankore on Aug. 6 in collaboration with Ooredoo Group, Nvidia, and Nokia. Ooredoo is the founding shareholder and lead investor, holding a 49% stake and committing approximately $800 million to the platform’s launch and initial development. That shareholder investment is separate from Zankore’s up-to-$3.1 billion bank loan facility.
Nvidia and Nokia are technology partners rather than identified equity owners in the cited disclosures. Nvidia is supplying accelerated computing technology, while Nokia is contributing networking infrastructure. Zankore’s planned 2027 capacity is expected to use Nvidia GB300 NVL72 systems alongside Nvidia’s DSX architecture for designing and operating AI factories.
The platform is intended to serve enterprises, startups, developers, and institutions across Indonesia and Southeast Asia, supporting AI training and inference workloads. Proceeds from the debt facility will fund the acquisition and deployment of Nvidia GPU infrastructure.
Zankore said its Nvidia collaboration also includes a revenue-sharing and credit-support model designed to align infrastructure deployment with customer demand, according to its Sept. 9 financing announcement. The company has not publicly disclosed anchor customers, contracted utilization, pricing, or key loan terms such as the interest rate and maturity.
Indonesia could capture more of Southeast Asia’s AI infrastructure
Zankore would add substantial accelerated-computing capacity inside Indonesia as Southeast Asian markets compete to host the infrastructure behind AI workloads. More domestic GPU capacity could reduce Indonesia’s reliance on compute hosted in neighboring markets while giving enterprises another regional option for training and inference.
The effects extend beyond servers. Large AI clusters require high-density data center space, high-capacity fiber, power generation and transmission, advanced cooling, and specialized operations talent. Those constraints are already reshaping AI data center development across APAC, where access to electricity and supporting infrastructure increasingly affects where new capacity can be built.
Indonesia could also capture more regional AI infrastructure investment that has historically concentrated in established hubs such as Singapore. Firmus Technologies is separately planning a 360 MW Nvidia-powered AI data center in Batam, whose proximity and connectivity to Singapore make the island a potential extension of the regional data center market while keeping more infrastructure investment inside Indonesia.
Zankore’s roughly 200 MW target for the first half of 2027 will be an early test of whether that shift gains scale. The loan gives the company substantial capital for GPU deployment, but reaching its broader regional 1 GW target will depend on customer demand, power availability, network capacity, and additional sites across Southeast Asia.
Let us teach you How to Talk to AI for free! Try our six-minute course at The Neuron Academy and learn a few simple ways to write better prompts and get more useful results from AI, or browse our other AI course for free for seven days. Check out all the lessons here →