Edgecore Networks has launched Praxis, a new Edge AI platform designed for managed service providers (MSPs), internet service providers (ISPs), and AI-powered SaaS and Video-as-a-Service (VaaS) companies looking to deploy artificial intelligence capabilities closer to customer environments.
The company said Praxis is aimed at technology providers that already operate AI-powered services in the cloud but are now seeking to expand those capabilities into large-scale edge deployments.
According to Edgecore, demand for edge AI is increasing across industries such as security, building management, retail, and industrial automation, where enterprises are increasingly expecting AI-enabled services as part of standard technology offerings.
The company said many AI service providers face challenges when deploying cloud-only AI architectures in real-world environments. These include integrating AI with existing infrastructure, controlling rising cloud inference costs, and meeting data privacy requirements for sensitive applications.
Edgecore said Praxis is designed to address those issues by enabling AI inference directly at the edge, where data is generated. The company noted that many enterprise customers continue to operate legacy infrastructure that was not originally designed for AI workloads.
Praxis allows customers to add AI capabilities without fully replacing existing systems, according to the company.
Edgecore also said moving AI inference from centralized cloud infrastructure to edge environments could help service providers reduce operational costs as AI adoption scales. Running inference entirely in the cloud creates increasing compute and bandwidth expenses, especially for video analytics and real-time applications.
The platform is also designed for industries where data sovereignty and privacy requirements prevent sensitive data from leaving on-premises environments. Edgecore said Praxis supports local data processing and storage for sectors such as healthcare, finance, government, and enterprise security.
The Praxis lineup includes five hardware configurations ranging from 1 TOPS to 70 TOPS of AI processing performance. The systems are powered by platforms from Synaptics and Qualcomm.
Edgecore said the platform supports a range of edge AI applications, including IoT connectivity, site-level video analytics, multi-stream AI inference, and enterprise-scale deployments.
The company added that each configuration has been validated for real-world edge AI deployments, allowing customers to select hardware based on application-specific requirements rather than overbuilding infrastructure.
Edgecore said Praxis builds on the company’s experience in enterprise and service provider networking infrastructure. The company has positioned the platform as a hardware foundation for AI service providers looking to accelerate deployment without building custom edge infrastructure from scratch.
Teng Tai Hsu said the shift toward edge AI is creating new opportunities for service providers seeking to deliver AI-enabled services closer to customers.
Edgecore Praxis is currently available for evaluation.
