on Posted on Reading Time: 3 minutes
At Mobile World Congress (MWC) 2025, MEF, Infosys, Nvidia, and IronYun together showcased a live demonstration of GPU-as-a-Service (GPUaaS)—a new approach to delivering AI at the Edge. By leveraging Graphics Processing Units (GPUs), service providers can offer scalable, low-latency AI services directly at the network’s edge, transforming how enterprises harness AI in real time.
MEF CEO Nan Chen and Infosys Vice President Jinu Koshy walked through a live demonstration of GPUaaS highlighting how telecom operators can seamlessly provision GPUs through MEF’s Lifecycle Service Orchestration (LSO) APIs, enabling customers to order resources on demand. This brings AI closer to where data is generated, enabling industries to benefit from faster, more efficient AI-powered solutions.
GPUs at the Edge: A Game Changer for AI
Traditional AI models rely on cloud-based computing, which introduces latency due to the distance between data sources and processing centers. For real-time AI applications like computer vision, traffic management, and security systems, it is essential to move AI processing to the network edge, where data is generated, latency is minimized, and performance is drastically improved.
The key to unlocking AI at the Edge lies in GPU-powered infrastructure. GPUs are ideal for AI inferencing tasks, enabling faster data processing for complex applications. The demonstration from MEF, Infosys, Nvidia and IronYun illustrates how high-performance GPUaaS makes it easier for enterprises to access powerful AI capabilities without heavy upfront investments.
AI in Action: Smart Cities at the Forefront
An example showcased at MWC was the use of AI in smart cities. With thousands of cameras generating massive streams of data, AI systems need the ability to process in real time. By using GPUs at the Edge, data can be analyzed on-site, enabling instant insights and actions like monitoring traffic flow, enhancing security, or optimizing energy use.
The application demonstrated how service providers can leverage MEF’s LSO APIs to automate provisioning of GPU service orders and AI models, simplifying service delivery and creating new revenue opportunities with AI-powered services at the Edge.
Standardization and Interoperability: MEF’s Role
The standardization and automation enabled by MEF’s LSO APIs ensures seamless, on-demand access to GPU resources across multiple platforms, providing an automated process for quoting, ordering, and activating GPU services. This creates a frictionless experience for both service providers and their customers, ensuring interoperability across different networks.
The ability to offer GPUaaS at the Edge is a transformative step in the evolution of AI. By utilizing MEF standards-based LSO APIs, service providers can offer enterprises a scalable, efficient way to deploy AI applications without significant capital investments, setting a new benchmark for AI services.
Empowering Enterprises with Real-World AI Solutions
Infosys, with its advanced solutions such as Infosys Topaz and Infosys Cobalt, is at the forefront of delivering high-performance AI at the Edge. By bringing AI processing to the Edge, Infosys enables enterprises to run real-time AI applications with lower latency and greater efficiency, while unlocking new revenue streams.
IronYun, a leader in video analytics, adds another layer of real-world applicability. Its Vaidio platform, powered by GPUs at the Edge, supports real-time video analytics for security, safety, and operational applications.
The Road Ahead: AI-Powered Networks at the Edge
The introduction of GPUaaS at the Edge is a major milestone in the evolution of intelligent telecom networks. As service providers adopt these AI technologies, they can unlock new monetization opportunities, offering high-value services while meeting the growing demand for low-latency solutions. As industries increasingly embrace AI, MEF, Infosys, Nvidia, and IronYun are leading the way in making these technologies accessible, scalable, and ready for widespread adoption, reshaping the landscape of AI and Edge computing for years to come.
Learn More
- See the recent MWC press release, MEF, Infosys, and IronYun Showcase NaaS GPU-as-a-Service for AI at the Edge Powered by NVIDIA.
- Also see the MWC press release, MEF Unveils NaaS Network APIs for an AI-Driven Economy and Application-Led Connectivity.
- Get the MEF NaaS Industry Blueprint.
- Get the MEF NaaS Customer Experience White Paper.
- Learn more about MEF LSO APIs.
- Meet Pascal Menezes at MEF’s next Global NaaS Event
Pascal Menezes
Chief Technology Officer | MEF