Significant Performance Gains Reinforce Imagination’s GPU-First Strategy

New data for E-Series show how one processor and one software stack addresses graphics, compute and AI workloads

London, UK – September 21, 2026 – Imagination Technologies (“Imagination”) today published the first performance data and demonstrations for its E-Series GPU IP and introduced Neural Super Resolution, a new AI-accelerated upscaling solution.  These updates reflect Imagination’s GPU-first strategy: providing chip designers with one architecture and software stack that can run graphics, compute and AI workloads.

As edge devices take on more demanding applications, chip designers are responding by adding new acceleration blocks, software stacks and data paths optimised for different workloads. This fragmentation adds integration effort and risks leaving fixed-function capacity unused as workload demand changes.  

More recently, different edge AI processors have begun to converge on a similar feature set, balancing performance and efficiency with flexibility. The GPU provides a natural starting point for such converged AI architectures, combining parallel acceleration, a proven software ecosystem, general-purpose flexibility – and graphics functionality.

E-Series, first announced last year integrates programmable AI acceleration efficiently alongside the GPU’s rendering pipelines. The result is one processor and one software stack that can run workloads as diverse as generative AI, neural rendering and gaming, either independently or alongside CPUs and other accelerators.

Programmable Acceleration for On-Device AI

AI models and their underlying operations continue to evolve long after a chip’s architecture has been defined. The general-purpose nature of E-Series, its familiar GPU-based programming model and use of industry-standard APIs allows developers to implement new operators throughout a device’s lifecycle. These are accelerated through the GPU’s tightly integrated Matrix Accelerator with support for both high- and low-precision operations (including BF16, FP4 and MX data formats).

The result is a processor that reaches 4.7x faster prefill performance for a typical edge language model than the previous D-Series generation. For Qwen 3.5 4B, a quad-core E-Series GPU at 1.5GHz can achieve time to first token of 0.2 seconds and decode throughput of 150 tokens per second on a representative workload*.

To simplify porting to Imagination GPUs, an optimised backend for Llama.cpp is being upstreamed, with native PyTorch and ONNX Runtime to follow.

Compute: A Strong Foundation for Intelligent Computing

Beyond headline applications, E-Series improves performance across the foundational compute operations used throughout AI, computer vision, signal processing and general-purpose computing.  For example, E-Series runs Conv2D kernels 4.4x faster and GEMM kernels 4.8x faster than a D-Series equivalent. GPU utilisation levels of 89% are possible for matrix multiplication workloads.

Neural Rendering: Bringing AI Into Graphics

Convergence creates new opportunities within graphics pipelines through neural rendering.  Imagination’s new Neural Super Resolution (NSR) technology applies E-Series’ matrix acceleration to the graphics pipeline using industry-standard extensions.

The temporal upscaling solution is deeply optimised for Imagination GPUs and combines a single-pass approach with Imagination’s proprietary self-compression solution which removes approximately 65% of the network’s weights. The resulting model is more power efficient than competing solutions and completes a typical 540p to 1080p upscale operation within as little as 2.3ms latency on a single-core E-Series GPU at 1GHz.

When compared to native rendering, NSR delivers a significant increase in frame rates while halving memory bandwidth consumption, all while delivering quality visually close to ground truth. NSR will be available as a library within the PowerVR SDK and through integrations for Unreal Engine and Godot.

Gaming: Advancing the Central Function of the GPU

Graphics remain central to E-Series. The architecture delivers significant graphics improvements, including support for DirectX 12 FL11_0. In real-world gaming workloads, E-Series delivers performance improvements of up to 54% compared with the previous generation equivalent and the quad-core configuration can scale to over 60fps for selected AAA desktop titles. Performance per watt improves by up to 39%.

Supporting Quotes

Jon Peddie “The boundaries between traditional processor categories are blurring. CPUs are using vector extensions to boost AI performance, while NPUs are becoming more general purpose. As functionality converges, Imagination is betting that the GPU is best placed to succeed; after all, it has the advantages of parallelism, an established software ecosystem and unlike any other processor it can run graphics in addition to AI and compute workloads.”

Markus Mosen, CEO of Imagination Technologies: “Edge workloads have diversified massively in the last decade, resulting in processor fragmentation, software complexity and underutilised silicon. Imagination’s GPU roadmap is designed to address such challenges. By consolidating graphics, compute and AI functionality onto one programmable architecture, we are providing chip designers with a processor worth building on.”

Heng Zhang, General Manager, XiangDiXian Computing Technology: “Imagination shares our ambition of bringing high-performance rendering to a broader range of devices. They have been our partner across multiple generations, thanks to their forward-looking GPU roadmap and a pragmatic approach to GPU design that prioritises real-world performance and user experience over peak TOPS and FLOPS.”

Clay John, Technical Director at W4 Games [Godot Engine]: “Developers usually have to make a trade-off between visual fidelity and performance.  Hardware-optimised upscaling solutions shift that balance and provide more headroom to improve visual fidelity at the same or better levels of performance.”

Explore Imagination IP:

• E-Series GPU IP
• PowerVR Series7XT GPU
• PowerVR Series9XM Graphic Processor
• Ray tracing GPU

E-Series Availability

E-Series configurations are available for applications scaling from smartphones up to high performance systems. Multiple lead partners have already licensed the technology, with first silicon expected to tape out later this year. To learn more about E-Series and explore detailed performance results, visit www.imaginationtech.com.

* Qwen 3.5 4B dense model, w4a16, on a quad-core EXD-64-2048 at 1.5 GHz; context length of 1,500 tokens and an output length of 1,500 tokens.