Activeport Expands into $25B AI Market with New GPU-Powered Model Routing Platform
Activeport Group Ltd has launched a new AI model routing platform that leverages its GPU orchestration expertise to deliver secure, sovereign AI hosting with hybrid cloud compatibility.
- Launch of AI model routing platform for private and public cloud AI execution
- Private GPU execution on sovereign, secure local clusters
- Hybrid cloud compatibility with major AI platforms like AWS Bedrock and Google Vertex
- Smart model routing based on cost, speed, and data residency
- Strategic expansion from cloud gaming GPU orchestration to AI inference
Activeport’s Strategic Shift into AI Inference
Activeport Group Ltd (ASX, ATV) has announced a significant expansion of its technology portfolio with the launch of its AI model routing platform. This new offering marks the company’s pivot from its established role in GPU orchestration for cloud gaming to becoming a key player in the rapidly growing AI inference market. By leveraging its deep expertise in GPU orchestration, Activeport aims to provide sovereign, secure, and low-latency AI hosting solutions tailored for telecommunications, government, and enterprise customers.
Sovereign AI Hosting with Hybrid Cloud Flexibility
The Activeport AI Platform delivers private GPU execution on local clusters, ensuring data sovereignty and compliance; a critical requirement for many organisations wary of public cloud constraints. The platform supports hybrid cloud deployments, seamlessly integrating with major public AI services such as AWS Bedrock, Google Vertex AI, Cerebras, and Groq. This flexibility allows customers to balance performance, cost, and regulatory demands while maintaining control over sensitive data.
Smart Model Routing and Scalable Orchestration
One of the platform’s standout features is its intelligent model routing capability. It dynamically selects AI models in real time based on factors like cost efficiency, processing speed, context window size, and data residency requirements. Coupled with scalable orchestration that auto-provisions GPU nodes and manages workloads with zero-touch scalability, Activeport’s solution promises to streamline AI inference operations at scale.
Building on Proven Telco-Grade Technology
Activeport’s AI platform builds on its existing telco-grade orchestration software, which currently powers massive GPU deployments for leading telecommunications companies’ cloud gaming services. This proven foundation gives Activeport a competitive edge in delivering reliable, enterprise-grade AI infrastructure. Chairman and CEO Peter Christie highlighted the natural progression from gaming to AI inference, noting particular excitement about hosting Diffusion models that benefit from the company’s streaming optimisation expertise.
Tapping into a $25 Billion Market Opportunity
The global AI inference market, valued at $25 billion, is increasingly shifting towards private, low-latency, and sovereign execution environments. Governments and enterprises are seeking alternatives to public APIs that often suffer from latency issues, compliance challenges, and unpredictable costs. Activeport’s AI Gateway leverages its GPU-as-a-Service technology to meet these demands, positioning the company as a preferred software layer for private AI clouds.
Bottom Line?
Activeport’s move into sovereign AI hosting could redefine private cloud inference, but market adoption and competitive dynamics will be key to watch.
Questions in the middle?
- Which early customers or sectors will lead adoption of Activeport’s AI platform?
- How will Activeport differentiate itself against major cloud AI providers?
- What financial impact and revenue growth can investors expect from this new AI focus?