About Open Parallels

Open Parallels is a blog dedicated to AIoT (Artificial Intelligence of Things) accelerators — the chips, boards, and platforms that bring machine learning to edge and IoT devices.

This site was developed to write about different AIoT accelerators and to present the results. Here I put each accelerator through real workloads and share what I find — benchmarks, throughput and latency, power efficiency, toolchain and developer experience, and practical notes from getting them running.

The goal is hands-on, comparative insight: helping engineers, researchers, and enthusiasts understand how these accelerators actually perform so they can choose the right one for their edge-AI projects.

Have a question or want to suggest an accelerator to test? Head over to the Contact page.