日本語 
CDNetworks, a leading edge services and network provider in APAC, announced on June 25, 2026 the availability of new delivery and security solutions for its multi-model AI platform, which integrates and delivers multiple generative AI models.
As the use of generative AI in enterprises increases, there is a rapidly growing demand for AI gateways, AI agents, multi-model AI platforms, and AI aggregation platforms that dynamically utilize multiple LLMs (Large Language Models) according to application, cost, and performance. These platforms are becoming the core of generative AI architectures, acting as a control layer that enables model selection, inference request routing, and cost optimization.
On the other hand, with the global expansion of services, new operational and security challenges are required, such as increased latency due to cross-region communication, handling sudden spikes in access, and cyberattacks targeting publicly accessible APIs.
Also Read: Kyndryl and Microsoft Expand Sovereignty Services, Boosting Secure Cloud Adoption for Regulated Industries
CDNetworks’ new solution integrates a global network infrastructure with edge-native security features to support fast and secure AI service delivery across users, applications, and AI models worldwide.
By leveraging CDNetworks‘ global network, which comprises over 3,000 Points of Presence (PoPs) worldwide, we route AI inference requests to the optimal path, minimizing network latency and ensuring high availability and stable service quality.
SOURCE: PRTimes
English