Edge Impulse

What it does

Edge Impulse is an MLOps platform for edge and physical AI. It supports MCUs, NPUs, CPUs, GPUs, gateways, sensors, cameras, and Docker containers. The platform covers end-to-end pipelines, security, versioning, and model monitoring.

Edge Impulse is a platform for building and running machine learning on edge devices. It supports MCUs, NPUs, CPUs, GPUs, gateways, sensors, cameras, and Docker containers. The platform covers end-to-end pipelines, security, versioning, and model monitoring. A Developer plan costs $0 per month and is aimed at individual developers, students, and universities exploring edge AI and prototyping. The Enterprise plan is for professional developers and teams that need to build datasets, train and optimize models, and deploy to edge devices. It uses custom pricing. The pricing page lists limits for the Developer plan, including up to 3 collaborators per project, 60 minutes of compute time per job, and 10 experiments per project. The EON Compiler is only available for models trained in Edge Impulse on the Developer plan, and excludes RAM Optimized mode. The Enterprise plan includes RAM Optimized mode and the compiler for all models, including models trained outside of Edge Impulse. The pricing page also lists API access as limited on Developer and full on Enterprise. The homepage does not state whether the software is open source. The homepage does not state whether an account is required, though it shows a login link and a Get started button.