A specialized processor designed to run AI and machine learning tasks efficiently, often built into phones and laptops.
A neural processing unit (NPU) is optimized for the mathematical operations neural networks rely on, especially large numbers of multiplications and additions on matrices. Compared with a general-purpose CPU, it can run these workloads faster and with far less power, which is important for battery-powered devices.
NPUs enable on-device AI features such as real-time photo enhancement, voice transcription, background blur on video calls, live translation, and running smaller language models locally. Keeping processing on the device can reduce latency and improve privacy, since data does not need to leave the phone or laptop. Similar accelerators in data centers include tensor processing units and other custom AI chips. Software support remains a practical factor in how useful an NPU is.
Central Processing Unit: the main processor in a computer that executes program instructions and coordinates other components.
GPUGraphics Processing Unit: a processor with thousands of small cores designed for parallel work, used for graphics and AI.
InferenceUsing a trained machine learning model to make predictions or generate outputs on new data.
On-Device AIAI that runs directly on a phone, laptop, or other device instead of sending data to remote cloud servers for processing.
QuantizationReducing the numerical precision of an AI model's weights, for example from 16-bit to 4-bit numbers, to make it smaller and faster.
Memory shortages, AI PCs, foldable iPhones and smart glasses. Here's which 2026 consumer tech shifts will last, which are hype, and what to buy or skip.
CPUs handle complex tasks one step at a time; GPUs crunch huge batches of simple math in parallel. Here's what that means for gaming, editing and AI.
AI is mostly matrix math, and GPUs do thousands of those calculations at once with very fast memory. Here's why, and what it means if you're buying a GPU for AI.