NPU (Neural Processing Unit) is a specialised accelerator embedded in phones, laptops and tablets, built to run neural-network workloads at very low power. Apple's Neural Engine, introduced with the iPhone X in 2017, brought the idea into the mainstream; Qualcomm, Samsung, Intel and AMD have since added their own NPUs to consumer silicon. With the rise of On-Device LLMs, NPUs have moved beyond face detection into running small language models, TTS and STT locally — Apple Intelligence and Microsoft's Copilot+ PC category both depend on this hardware. NPUs are less flexible than GPUs for general workloads, but they are best-in-class for energy efficiency at sustained AI inference.
MEVZU N°124ISTANBULYEAR I — VOL. III
Glossary · Intermediate · 2017
NPU (Neural Processing Unit)
A specialised AI accelerator integrated into phones and laptops to run neural workloads efficiently.
- EN — English term
- NPU (Neural Processing Unit)
- TR — Turkish term
- NPU — Nöral İşlem Birimi