Edge AI Inference
Running smaller models directly on-device or at the network edge.
Falling model sizes and better on-device accelerators make edge inference viable for latency-sensitive or offline-first product experiences — worth prototyping, not yet worth betting the architecture on.
Sequencing
Builds on Small Language Models (SLMs)
Edge AI Inference across other industries
Running smaller models directly on-device or at the network edge.
Edge AI for Bedside & Wearable Devices
Inference running directly on monitoring and wearable devices.
Edge AI for Branch & ATM Devices
On-device inference for branch and self-service channel devices.
Edge AI on the Factory Floor
Inference running directly on or near production-line hardware.
Edge AI for In-Store & POS Devices
Inference running directly on point-of-sale and in-store hardware.
From the Technology & SaaS Emerging Technology & Innovation Radar, Edition 2026.1, published 2026-10-01.
An independent, industry-agnostic radar curated by Shailesh Kumar.