New AI Solution Cuts Costs by 86%, Boosts Throughput by 3x
Akamai has launched Akamai Cloud Inference, a new AI service designed to make large language models (LLMs) and predictive AI faster, more efficient, and cost-effective.
“Inference is the next frontier for AI,” said Adam Karon, Chief Operating Officer and General Manager, Cloud Technology Group at Akamai. “Businesses need AI to work in real-time, close to users. That’s where Akamai’s globally distributed platform makes all the difference.”
Akamai Cloud Inference promises up to 86% lower costs, 3x better throughput, and 2.5x lower latency compared to traditional hyperscale cloud infrastructure. The service runs on Akamai Cloud, the world’s most distributed cloud platform, designed to bring AI inference closer to users by leveraging 4,200+ points of presence across 130+ countries.
Powering AI at the Edge
Akamai Cloud Inference provides:
- Optimized Compute: CPUs, GPUs, and AI-specific VPUs with NVIDIA AI Enterprise integration.
- Advanced Data Management: Real-time data access powered by VAST Data and vector database support for retrieval-augmented generation (RAG).
- Containerization & Edge Compute: Kubernetes-powered deployment and WebAssembly (Wasm) integration for low-latency AI at the edge.
With AI shifting from training massive models to real-time inference, Akamai Cloud Inference is already being used for in-car voice assistants, AI-driven crop management, e-commerce optimization, and real-time sentiment analysis.