Akamai Technologies has announced a major leap in AI infrastructure with the launch of its AI Grid intelligent orchestration platform, enabling distributed AI inference across more than 4,400 edge locations worldwide.
Built as part of the Akamai Inference Cloud, the solution represents the industry’s first global-scale implementation of NVIDIA AI Grid architecture, designed to move AI workloads beyond centralized data centers to a distributed, edge-to-core compute model.
The platform integrates thousands of NVIDIA RTX PRO 6000 Blackwell GPUs, allowing enterprises to run real-time AI applications—ranging from generative AI to physical and agentic AI—with the speed of local compute and the scale of global infrastructure.
At the core of the AI Grid is an intelligent orchestration engine that dynamically routes AI workloads across edge, regional, and core environments. This enables enterprises to optimize “tokenomics” by balancing cost per token, latency, and throughput, ensuring workloads are executed on the most efficient compute tier.
“Real-time AI applications require inference at the point of interaction, not centralized clusters,” said Adam Karon, COO and GM, Cloud Technology Group at Akamai. “Our AI Grid extends AI factories outward, using a distributed architecture to deliver performance, cost efficiency, and scalability.”
Driving Real-Time AI at Scale
The platform is designed to support a wide range of enterprise use cases:
- Gaming: Sub-50 millisecond AI-driven NPC interactions and immersive gameplay
- Financial Services: Real-time fraud detection and hyper-personalized customer experiences
- Media & Entertainment: AI-powered video transcoding and real-time dubbing
- Retail: In-store AI applications and intelligent customer engagement at the point of sale
Akamai’s architecture combines edge computing, high-performance GPU clusters, and intelligent routing, enabling real-time responsiveness while reducing reliance on centralized infrastructure.
Edge-to-Core AI Infrastructure
Powered by NVIDIA’s Blackwell architecture and AI Enterprise stack, the platform spans:
- Edge (4,400+ locations): Low-latency processing for real-time AI interactions
- Core cloud & GPU clusters: High-density compute for large-scale model training and inference
“AI-native applications demand predictable latency and cost efficiency at scale,” said Chris Penrose, Global VP, Business Development at NVIDIA. “Akamai’s implementation of AI Grid is enabling the next wave of real-time, distributed AI applications.”
Redefining AI Infrastructure
With enterprises increasingly shifting toward inference-heavy workloads, Akamai’s distributed approach addresses the limitations of centralized AI models. By bringing compute closer to users, the platform enhances performance, reduces latency, and improves ROI for AI deployments.
The Akamai Inference Cloud is now available for enterprise customers, with early adoption already underway across industries requiring high-performance, low-latency AI capabilities.
This launch signals a broader industry shift — from centralized AI “factories” to globally distributed AI infrastructure, setting the stage for the next phase of scalable, real-time AI innovation.