Singtel's RE:AI Recognized by Frost & Sullivan for GPU-as-a-Service Strategy Leadership in Southeast Asia
Frost & Sullivan has awarded Singtel's RE:AI the 2025 Southeast Asia Competitive Strategy Leadership Recognition for its approach to GPU-as-a-Service (GPUaaS). For executives, the signal is clear: integrated AI infrastructure paired with sovereign-first controls is moving from vision to execution at regional scale.
Why this matters for enterprise strategy
- Integrated stack: Compute, connectivity, applications, and orchestration delivered as one service through Singtel's Paragon platform-reducing integration risk and time-to-value.
- Sovereignty and compliance: Full-stack control across data centers, network, GPUs, and orchestration to support regulated workloads in government, finance, healthcare, and research.
- Performance roadmap: On-demand access to NVIDIA H100s and GB200 today, with plans to introduce GB300 Grace Blackwell Superchips in 2026 for next-step model performance.
- Regional scale: Expansion to 200 MW by 2026 across Singapore, Malaysia, Thailand, and Indonesia to improve latency, capacity planning, and data residency choices.
- Multi-cloud optionality: Partnerships with Vultr, Nscale, GMI Cloud, and major hyperscalers for workload portability and vendor diversification.
- Operational efficiency: Liquid cooling, energy-efficient designs, and automation to lower energy use and support corporate sustainability targets.
"A holistic approach that combines innovation, scalability and sovereign-first infrastructure gives RE:AI a unique competitive advantage in Southeast Asia's AI landscape," said Renato Pasquini, Vice President, Research at Frost & Sullivan. "By leveraging its extensive data center assets, advanced connectivity and patented Paragon platform, RE:AI delivers GPU-as-a-Service solutions that meet current AI demands and anticipate future enterprise requirements."
Bill Chang, Chief Executive Officer, Digital InfraCo, Singtel, added: "We created RE:AI with the mission of enabling sovereign AI use cases by making AI open, accessible, secured and scalable. The capabilities are anchored by our patented Paragon orchestration platform and hyperconnected, sustainable AI-ready data centres... We're pleased that the unique competitive advantages of our RE:AI cloud service and Paragon platform have been recognised by Frost & Sullivan as we expand RE:AI into the region with our diverse partner ecosystem."
What's under the hood
- Paragon orchestration: Unified control of compute, storage, networking, and applications for consistent governance and lifecycle management.
- Telecom-grade connectivity: Private, low-latency links from network edge to GPU clusters to stabilize training and inference throughput.
- AI-ready data centers: Liquid cooling, renewable energy initiatives, and energy-aware scheduling to improve PUE and support net-zero commitments.
- Predictive operations: AI-driven monitoring across each GPU and data hall to sustain uptime and performance with minimal intervention.
- Customer success: Feedback loops and advisory support to help teams operationalize AI across use cases and regions.
Footprint and growth
Singtel's first-mover push in Singapore's AI infrastructure includes DC Tuas, a 58 MW site under development. The portfolio expansion aims to nearly double overall capacity to 200 MW by 2026 across Southeast Asia-giving enterprises more choices on proximity, price-performance, and compliance.
The Frost & Sullivan signal
The recognition follows Frost & Sullivan's benchmarking across two dimensions: strategy effectiveness and strategy execution. Their Best Practices recognitions are based on interviews, analysis, and secondary research to find providers improving market share, customer satisfaction, and competitive positioning.
Learn more about Frost & Sullivan's Best Practices framework here.
Executive decision checklist
- Use case fit: Do your most sensitive workloads (financial risk models, clinical analytics, public sector projects) require sovereign-first controls?
- Deployment model: Will you run across multi-cloud and need consistent policy, cost, and performance management?
- Data residency and compliance: Can you meet sector rules without sacrificing GPU availability and throughput?
- Performance and economics: Are SLAs, scaling behavior, and energy efficiency clear enough for board-level cost scrutiny?
- Roadmap confidence: Does the 2026 upgrade path (GB300) align with your model and data growth plans?
For leadership teams building organization-wide AI capability, explore practical learning paths by role at Complete AI Training.
Your membership also unlocks:
