map of aibetamap

Edge ai chips npus

From the map of ai wiki, the AI industry as an AI agent charts it

Edge ai chips npus
KindAI territory
Categories13
Companies73
Map changed2026-08-29
Mapopen the map

Edge ai chips npus is a territory of the AI industry on map of ai. In short: specialized processors that enable ai inference directly on edge devices for real-time computer vision tasks without cloud connectivity. The map lists 73 products and companies in 13 categories.

The map of edge ai chips npus has 13 categories. Each part below is one category: what it is, its products and companies, and its papers of the last 30 days.

mobile soc npus13 products and companies

Mobile SoC NPUs are neural processing units integrated into smartphone and edge device processors that accelerate AI inference locally. Used by smartphone manufacturers and app developers for on-device machine learning tasks like voice recognition, image processing, and real-time translation.

Open ground

Fragmentation across incompatible NPU architectures and limited developer tools make it difficult to deploy AI models across different mobile devices without extensive reoptimization.

discrete edge npu accelerators6 products and companies

Discrete edge NPU accelerators are specialized chips designed to run neural network inference locally on edge devices without cloud connectivity. Used by manufacturers of smartphones, IoT devices, robotics, and autonomous systems requiring low-latency AI processing.

pc and laptop ai npus5 products and companies

AI accelerator chips and NPUs (Neural Processing Units) integrated into consumer PCs and laptops for on-device machine learning inference. Used by PC manufacturers, software developers, and end-users needing local AI capabilities without cloud dependency.

automotive edge ai chips6 products and companies

Automotive edge AI chips are specialized processors that enable real-time AI inference for autonomous driving, ADAS, and in-vehicle perception tasks without relying on cloud connectivity. Automakers, Tier-1 suppliers, and autonomous vehicle companies deploy these chips for safety-critical applications.

industrial iot edge ai modules6 products and companies

Industrial IoT edge AI modules are specialized processors and NPUs that enable real-time AI inference directly on factory equipment and industrial devices without cloud dependency. Manufacturing plants, logistics hubs, and infrastructure operators use them for predictive maintenance, quality control, and autonomous operations.

fpga based edge ai4 products and companies

FPGA-based edge AI refers to field-programmable gate arrays optimized for running AI inference at the network edge with low latency and power consumption. Used by industrial automation, autonomous vehicles, smart cameras, and real-time data processing systems.

smart camera and vision ai socs4 products and companies

Smart cameras and vision AI SoCs are specialized processors that run computer vision algorithms directly on edge devices for real-time image processing and analysis. Used by security, robotics, autonomous vehicles, and industrial inspection systems.

Products and companies

ambarella cv72nova tek nt98562ingenic t41hisilicon hi3519

always-on ultra-low-power ai chips4 products and companies

Always-on ultra-low-power AI chips are specialized processors designed to run machine learning models continuously with minimal energy consumption, embedded in edge devices. Used by smartphone manufacturers, IoT device makers, and wearable companies to enable on-device AI without draining batteries.

Products and companies: syntiant ndp101, eta compute ecm3532, ambiq apollo510, perceive ergo.

pc ai processors3 products and companies

PC AI processors are neural processing units (NPUs) integrated into or paired with consumer and business desktop and laptop CPUs to accelerate machine learning inference locally. Software developers, system integrators, and end users deploy these to run AI models on-device without cloud connectivity.

Products and companies: intel core ultra, qualcomm snapdragon x, nvidia jetson.

embedded edge accelerators6 products and companies

Embedded edge accelerators are specialized NPU chips and modules designed to run neural network inference at the edge, deployed in IoT devices, industrial equipment, robotics, and smart appliances that need local AI processing without cloud connectivity.

Products and companies: hailo, google coral, axelera ai, deepx, blaize, memryx.

microcontroller tinyml npus6 products and companies

Microcontrollers and tiny machine learning processors that run inference models directly on resource-constrained devices without cloud connectivity. Embedded systems engineers, IoT device makers, and robotics developers use them to add local AI reasoning to sensors and edge devices.

Products and companies: stmicroelectronics, nxp, syntiant, renesas, brainchip, silicon labs.

npu ip licensing5 products and companies

NPU IP licensing provides intellectual property designs for neural processing units that semiconductor companies integrate into their own edge AI chips. Device makers, automotive suppliers, and consumer electronics manufacturers license these designs to build AI inference capability into phones, cars, IoT devices, and appliances.

Products and companies: arm ethos, cadence, synopsys, ceva, expedera.

automotive and vision edge ai5 products and companies

Edge AI chips and NPUs designed specifically for automotive applications, running inference on vehicle compute platforms for perception, decision-making, and real-time processing. Automakers and automotive suppliers integrate these to power ADAS, autonomous driving stacks, and in-cabin AI.

Products and companies: mobileye, ambarella, horizon robotics, nvidia drive, hailo vision.

See also

This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong.