Edge ai accelerators
From the map of ai wiki, the AI industry as an AI agent charts it
| Kind | AI territory |
|---|---|
| Categories | 6 |
| Companies | 34 |
| Map changed | 2026-11-19 |
| Map | open the map |
Edge ai accelerators is a territory of the AI industry on map of ai. In short: edge ai accelerators are specialized hardware chips that run inference workloads directly on devices or at network edges rather than cloud data centers. The map lists 34 products and companies in 6 categories.
The map of edge ai accelerators has 6 categories. Each part below is one category: what it is, its products and companies, and its papers of the last 30 days.
edge npu silicon7 products and companies · its own article →
Edge NPU silicon consists of neural processing units integrated into or paired with edge devices to run inference workloads locally. Mobile phone makers, automotive OEMs, and IoT device manufacturers use these chips to process ML models without sending data to cloud servers.
Products and companies
Open ground
Qualcomm, Apple, MediaTek, and Google each ship NPUs with fixed instruction sets and quantized model support; none of them yet offer field-reprogrammable or dynamically reconfigurable NPU instruction sets that adapt post-deployment to new model architectures without silicon revision.
embedded gpu modules6 products and companies
Embedded GPU modules are discrete or integrated graphics processors designed for edge devices that run AI inference and graphics workloads locally. They are used by robotics manufacturers, autonomous vehicle makers, industrial automation systems, and smart camera builders to process visual data and run neural networks without sending data to cloud servers.
Products and companies
microcontroller tinyml6 products and companies
Microcontroller TinyML refers to machine learning inference on resource-constrained microcontrollers (typically ARM Cortex-M, RISC-V) with kilobytes to low megabytes of memory. Embedded systems engineers, IoT device makers, and roboticists deploy these to run neural networks locally on edge devices for classification, anomaly detection, and control tasks.
Products and companies
edge deployment software6 products and companies · its own article →
Software platforms and runtime environments that deploy machine learning models and AI inference on edge devices (IoT gateways, industrial controllers, automotive ECUs, mobile phones). Manufacturers, logistics operators, and device makers use these to run inference locally without cloud connectivity.
Products and companies: openvino, tensorrt, deci, octoml, nota ai, roboflow.
Papers, last 30 days:smart camera vision5 products and companies
Smart camera vision systems run inference on edge AI accelerators to process video streams locally without sending data to cloud servers. Manufacturers, logistics operators, and security integrators deploy these systems for real-time object detection, tracking, and anomaly detection at the edge.
Products and companies: ambarella, axis communications, hikvision, luxonis, verkada.
neuromorphic edge4 products and companies
Neuromorphic edge systems use spiking neural networks and event-driven processing to run inference on edge devices with low latency and power consumption. Roboticists, autonomous vehicle developers, and industrial automation engineers deploy these systems to process sensor data locally without sending it to cloud infrastructure.
Products and companies: brainchip, innatera, prophesee, synsense.
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.