#USA #Zettabyte #AI Computing #Taipei City #zCLOUD

Zettabyte Launches Model-as-a-Service on zCLOUD

On August 12, 2026, Zettabyte, a prominent player in global AI computing, rolled out its newest offering, Model-as-a-Service (MaaS) on the zCLOUD platform. This innovative service allows teams to easily access and utilize several open-source AI models via an API, thus streamlining the development process in AI projects.

What is Model-as-a-Service?

The Model-as-a-Service provided by zCLOUD hosts a variety of widely used AI models, including Kimi, GLM, Llama, GPT-OSS, Gemma, and Qwen. With this structure, teams no longer need to undergo the cumbersome process of sourcing, installing, and tuning models on their own. Instead, they can immediately start building applications by calling these pre-configured models through the API powered by Zettabyte’s high-performance GPUs.

GPU Availability and Performance

Zettabyte offers access to a range of cutting-edge NVIDIA GPUs, including the H100, H200, B200, and B300. Users can take advantage of the performance these GPUs deliver, with the pricing structure beginning at an affordable US$1.99 per GPU-hour. This flexible pricing model adjusts according to specific requirements and scale, providing teams with an economical choice in hardware procurement.

The infrastructure behind zCLOUD is designed for optimal reliability, being backed by a 99.5% uptime SLA target, which is essential for enterprises relying on consistent performance in their AI operations. Additionally, reserved GPU clusters can be deployed swiftly, often within six hours, minimizing downtime for organizations that need immediate resources.

Flexible Access Options

Zettabyte has developed three distinct avenues for users to access GPU capacity:

• - Short-Term Capacity: This option allows for scaling GPU instances to hundreds without long-term commitments, ideal for teams with fluctuating needs.
• - Reserved Clusters: This service provides predictable and fully-managed capacity for continuous training and inference under a set term, perfect for larger projects that require stability over time.
• - Private Cloud: This customizable deployment caters to teams with specific procurement, security, or regional needs, offering tailored solutions that ensure compliance and alignment with organizational policies.

Bridging AI Compute Demand

Dr. David Ku, Technology Leader at Zettabyte, underscored the significance of this launch by saying, "AI computing will be everywhere, from hyperscale data centers to the edge. The supporting AI infrastructure and software must be compatible, resilient, and agile. zCLOUD was constructed to deliver Model-as-a-Service, bringing more than 40 hardware providers together and expertly managing the underlying AI infrastructure through our zSUITE full stack."

The introduction of the Model-as-a-Service on zCLOUD is now available to teams ready to innovate and accelerate their AI projects. For further details, potential users can explore the offerings at www.zettabytecloud.com and evaluate how this powerful service can enhance their AI computing capabilities.

About Zettabyte

Zettabyte operates at the forefront of AI computing globally, focusing on high-quality computation and efficiency. With its zSUITE platform, Zettabyte aims to empower organizations by enhancing reliability, readiness, observability, and energy efficiency in AI infrastructure. To learn more about Zettabyte and its innovative solutions, visit www.zettabyte.space.