Imagination Technologies announced the launch of its E-Series GPU IP, a next-generation architecture that redefines the role of GPUs in edge computing. Designed to meet the growing demand for high-performance, power-efficient AI and graphics processing, the E-Series positions the GPU as the central accelerator for both compute and visual workloads at the edge.
The E-Series introduces a scalable architecture that delivers up to 200 TOPS (INT8/FP8) of AI performance, while maintaining the advanced graphics capabilities Imagination is known for. This leap in performance is powered by two key innovations: Neural Cores and Burst Processors.
On-device AI is evolving rapidly, but edge AI system designers still face challenges in balancing performance and efficiency with flexibility,” said Phil Solis, Research Director at IDC. “Imagination has leveraged its long-held experience developing power-efficient GPUs and evolved them to flexibly support both graphics and AI workloads for on-device AI.”Burst Processors: A novel architectural enhancement that improves average power efficiency by 35%, ideal for sustained edge workloads such as computer vision, natural language processing, and user interfaces.
The E-Series is designed with future-proofing in mind. Its GPU programmability allows developers to adapt to evolving AI and graphics workloads using open standards such as OpenCL, oneAPI, Apache TVM, and LiteRT. Imagination’s own compute libraries and graph compiler further optimize performance and developer productivity.
“Edge AI hardware and software need to converge to unlock the full potential of on-device intelligence,” said Parv Sharma, Senior Analyst at Counterpoint Research. “E-Series gives compute software developers across every market the freedom to deploy the latest algorithms on any device.”
