Thaki Cloud Unveils ‘Thaki NeoCloud OS’

• Transforms existing AI infrastructure, including GPUs and AI Factories, into branded public cloud services.
• Enables telecom operators, data center providers, AI Factory operators, and GPU infrastructure providers to build and operate their own NeoCloud services.
• Integrates essential cloud operations capabilities, including provisioning, multi-tenancy, GPU scheduling, usage metering, billing, SLA management, and observability.
• Turns complex AI infrastructure into self-service, on-demand cloud services that customers can easily access and use.

SEOUL, South Korea, October 13, 2026 — Thaki Cloud, an enterprise AI infrastructure and platform company, today unveiled Thaki NeoCloud OS, a cloud platform designed to transform AI infrastructure, including GPU servers, storage, and networking, into public cloud services that customers can readily access and use.

Thaki NeoCloud OS enables organizations that own GPU and AI infrastructure—including telecommunications companies, data center operators, AI Factory operators, and GPU infrastructure providers—to build and operate their own branded NeoCloud services.

As enterprises and infrastructure providers continue to invest in large-scale AI infrastructure, including GPU clusters and AI Factories, a critical distinction is emerging: building GPU infrastructure is not the same as delivering it as a cloud service.

Commercializing GPU infrastructure as a cloud service requires an additional operational layer encompassing customer and tenant management, resource provisioning, GPU scheduling, network and storage configuration, usage metering, billing, service-level agreements (SLAs), monitoring, incident response, and resource reclamation.

Thaki Cloud developed Thaki NeoCloud OS to bridge this gap between physical AI infrastructure and commercially operational cloud services.

From Owning GPU Infrastructure to Delivering Cloud Services

The purpose of Thaki NeoCloud OS extends beyond simply managing GPU resources. Its primary function is to transform physical AI infrastructure—including GPU servers, storage, and networking—into cloud service offerings that customers can select, provision, and use on demand.

Thaki NeoCloud OS brings together the essential capabilities required to operate a cloud business within a unified cloud operations platform. These include service catalogs, user and organization management, multi-tenancy, GPU and bare-metal provisioning, GPU scheduling and resource allocation, automated network configuration, storage integration and provisioning, quota and policy management, usage metering, billing and settlement integration, SLA management, monitoring and observability, capacity management, resource lifecycle automation, and API-based integration with external systems.

With these capabilities, infrastructure operators can securely isolate GPU resources for individual customers, automatically allocate resources in response to service requests, measure consumption, and manage service levels and resource lifecycles.

Making Complex AI Infrastructure Easy to Consume

Thaki NeoCloud OS is designed not only to support infrastructure operators but also to simplify the experience for end users.

Traditionally, accessing GPU infrastructure has required multiple configuration steps, including server allocation, network setup, storage configuration, operating system installation, and preparation of execution environments.

Thaki NeoCloud OS abstracts much of this complexity within the cloud platform, allowing users to select and rapidly access the GPU and computing resources they need.

This enables infrastructure providers to expand beyond conventional GPU server rental models toward a more comprehensive business model: Infrastructure → Product → On-Demand Cloud Service.

A key objective is to deliver a self-service experience in which AI developers and enterprise customers can select and consume computing resources without needing to understand or manage the underlying infrastructure complexity.

A Platform for Building Your Own NeoCloud

Thaki NeoCloud OS is not a single cloud service operated directly by Thaki Cloud. Instead, it is a cloud platform that enables infrastructure owners to build and operate their own branded NeoCloud services using their existing GPUs, servers, storage, networking, and data center infrastructure.

Telecommunications companies can leverage their GPU infrastructure and data centers to establish their own AI cloud services.

Internet data center (IDC) and data center operators can expand from traditional GPU hosting into on-demand GPU cloud services.

AI Factory operators can transform their existing facilities into cloud services capable of serving multiple customers.

GPU infrastructure providers can turn their computing capacity into cloud offerings accessible through APIs and self-service interfaces.

Each operator retains control over its brand, service policies, pricing models, and infrastructure configuration, while Thaki NeoCloud OS provides the underlying cloud platform needed to deliver and operate those services.

The Next Step for AI Factories Is Cloud Services

As investment in large-scale AI infrastructure and AI Factories continues to grow, connecting newly deployed GPU resources to actual services and sustainable business operations is becoming an increasingly important challenge.

Building the infrastructure is only the beginning. To make those resources commercially accessible, operators also need a cloud operations framework that allows customers to consume them as services.

Transforming an AI Factory into a cloud service involves three fundamental layers:

1. Infrastructure — GPU / Servers / Networking / Storage / Data Centers
2. Thaki NeoCloud OS — Provisioning / Multi-Tenancy / Scheduling / Usage Metering / Billing / SLA Management / Observability / Lifecycle Management
3. Customer NeoCloud Service — Web Console / APIs / Self-Service / On-Demand GPU / AI Workloads

Thaki NeoCloud OS serves as the software layer connecting physical infrastructure to an operational cloud business.

With this platform, AI Factory operators can move beyond infrastructure deployment and develop cloud businesses that continuously deliver computing resources to customers, track consumption, and support usage-based billing.

Kang So-ryun, CEO of Thaki Cloud, commented:

“Securing GPUs and building an AI Factory are fundamentally different from creating a cloud service. For customers to actually use these resources, operators need a cloud platform that can turn infrastructure into commercial offerings, automatically provision resources in response to orders, measure usage, and enable billing.”

He added:

“Thaki NeoCloud OS is designed to help telecommunications companies, data center operators, AI Factory providers, and other infrastructure owners build and operate their own NeoCloud businesses using the AI infrastructure they already possess. Our goal is to transform complex GPU infrastructure into cloud services that customers can access and use easily and quickly.”