Imagination Technologies is unveiling its E-Series graphics processing units (GPUs) for graphics and AI processing at the edge.
Imagination said the products redefine edge AI and graphics system design with the launch, leveraging a highly efficient parallel processing architecture to provide exceptional graphics performance while also scaling from two to 200 TOPS INT8/FP8 for AI workloads.
In a briefing, Kristof Beets, vice president of product management at Imagination, told me that the GPUs have a clever blending of AI processing components within the chip so that it most efficiently handles the processing tasks at hand. That helps distinguish it from other GPUs on the market and enables Imagination to target specialty markets like automotive.
Imagination says its GPU design is unique when it comes to AI acceleration.
He said the GPU family offers a versatile and programmable solution for future edge applications including graphics, desktop applications, natural language processing on smartphones, industrial computer vision, and vehicle autonomy.
Two new technologies underpin E-Series’ potential to transform edge system design:
Neural Cores: Scaling up to 200 TOPS (INT8/FP8), these cores deliver significant acceleration for AI and compute workloads.
Burst Processors: A highly innovative solution delivering a 35% improvement in average power efficiency for edge applications.
“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, in a statement. “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. The E-Series combines the programmability of GPUs with a boost in AI performance to create a compelling solution for on-device AI system designers.”
High performance acceleration for low power AI
Imagination Technologies’ E-Series GPUs are targeted at edge AI.
E-Series continues to offer the advanced graphics capabilities of previous generations of Imagination GPUs, including support for ray tracing. To this, it adds deeply integrated acceleration for power-efficient, low precision AI operations into every GPU core. This creates the compute-dense E-Series Neural Cores that scale up to 200 TOPS INT8 and unleash up to 400% the AI performance of the previous D-Series.
The Neural Cores support a wide range of popular AI number formats, enabling developers to design networks that meet a broad spectrum of performance, accuracy and power demands. One of their many performance efficiency measures is an AI-friendly memory architecture that prioritizes local memory for compute, greatly reducing the power and performance costs of going to external memory.
Programmable AI for future-proof system design
Imagination Technologies says its AI design is better.
GPUs are pro …