A New Era Of Edge AI: E-Series GPU IP

Welcome to the Age of Parallel Compute

The age of parallel compute has arrived. The emergence of artificial intelligence is having profound implications on hardware development in every single market. Data centres are being built with massive compute resources to handle the demands of training and connected inference. At the edge, embedded hardware system designers are dealing with the challenges of delivering AI in power, memory and compute constrained environments.

AI at the Edge is essential for applications where connectivity may be limited, latency isn’t tolerated, privacy is needed or power constraints require efficient local processing. However, for it to be a success, system designers need to unlock a significant uplift in compute performance within the well-established power, memory and area constraints of edge devices.

Extracting maximum performance from artificial intelligence algorithms is, at its core, a parallelisation problem. The computational tasks within an AI model demonstrate independence (they do not rely in intermediate results from other tasks), require minimal data exchange during execution and are decomposable (processing can be split into a single set of many identical tasks). They can be parallelised into hundreds or thousands of simultaneous operations in order to boost performance.

Ultra-parallel workloads are exactly what GPUs are designed to handle. AI algorithms have a lot of similarities to 3D graphics, where each frame can be decomposed into pixels for independent, parallel processing. Imagination has been developing its energy-efficient parallel processing technology for graphics acceleration for over three decades, and our GPUs are now a natural fit for AI workloads of the future, particularly at the edge. In 2025, Imagination introduced its E-Series GPU IP: an exceptionally efficient parallel compute engine for on-device AI and graphics processing with the ability to scale up to 200 INT8 TOPS, E-Series puts the GPU at the centre of AI at the edge. This paper outlines the technology that will feature across all E-Series GPUs.

Read more here.