Tiny GPU without compromise
Based on a highly power-conscious architecture, our IMG A-Series products are perfect for industrial and wearable applications, with higher performance configurations available for mainstream mobile.
IMG A-Series supports the latest APIs including OpenGL ES, Vulkan and OpenCL. Software optimisation tools enable developers to quickly analyse frames and identify problem areas.
This 128-wide arithmetic logic unit (ALU) design, simplified compared to previous generation architectures, enables greater parallelism for up to four times more operations per clock. Compiler optimisations ensure maximum utilisation resulting in a massive increase in performance per millimetre and greatly improved power efficiency.
AR / VR
High frame rates and engaging immersive experiences across gaming, socialising, exploring and business applications.
Entry level DTV
Smooth graphical user interfaces for 1080p screens that make choosing what show to watch a pleasure.
Efficient and safe computing for communication networks, safe cities and other industrial applications.
Set top box
Highly efficient content streaming experiences for screens from 1080p to beyond 4K.
Smart home hub
Speakers combined with inbuilt screens transform the home experience by adding a smart way to interact with digital life.
Ultra small, ultra low power screens whose detailed user interfaces still keep up with the touch of a finger.
Full Features & Specs
- Vulkan 1.3
- OpenGL ES 3.x/2.0/1.1 + Extensions
- OpenCL 3.0
- Android NN HAL
- Linux Consumer
- Linux X.org
- PVRTC and ETC Support
Floating point operations per clock
- 16 FP32 FLOPs/Clock
- 32 FP16 FLOPs/Clock
Pixels per Clock
HyperLane Technology overview
Download the HyperLane overview to find out how our newest technology can give you the edge.Read more
IMG A-Series: A True Performer
Imagination’s lowest area GPU IPRead more
HyperLane: PowerVR Virtualization Explained
An overview of GPU virtualization and its associated challenges and highlights the unique features of Imagination’s GPUs.Read more