Glossary
Plain definitions of the terms behind AI agents that operate phones and apps.
Agents that use apps
Mobile use
Mobile use is an AI agent's ability to operate the apps on a phone or tablet and finish tasks for a person.
Phone use
Phone use is an AI agent using the apps on a phone to finish tasks, often on the person's own phone and accounts.
Agentic OS
An agentic OS is an operating system with an AI agent that finishes tasks across the device's apps when the user asks.
Computer use
Computer use is an AI agent operating a computer's apps and websites through the screen, the way a person does.
Agentic commerce
Agentic commerce is shopping that an AI agent completes for a person, from finding the item to checkout, with the person approving the payment.
Agent layer
An agent layer is the part of a device's software that turns a person's request into finished tasks across the device's apps.
Agent harness
An agent harness is the software around an AI model that runs a task from start to finish and keeps the person in control.
Accessibility service (Android)
An accessibility service is an Android feature that lets an app you approve read the screen and act in other apps on your behalf.
MCP (Model Context Protocol)
MCP (Model Context Protocol) is an open standard that lets AI assistants and agents connect to outside tools and data.
Models
Action model
An action model is an AI model that returns the next action in an app or on a screen instead of a written answer.
On-device action model
An on-device action model is an action model small enough to run on a phone's or PC's own chip.
Small action model (SAM)
A small action model (SAM) is an action model compact enough to run on a phone, PC or other edge device.
Vision-language-action (VLA) model
A vision-language-action (VLA) model takes an image and an instruction and returns an action, for a robot or on a screen.
On-device AI
On-device inference
On-device inference means running a trained AI model on the device itself, such as a phone or PC, instead of on a server.
Hybrid inference
Hybrid inference runs an AI model on the device where the hardware allows and sends heavier steps to the cloud or on-prem servers.
Inference engine
An inference engine is the software that runs a trained AI model on the chips a device has.
NPU (neural processing unit)
An NPU (neural processing unit) is the part of a chip built to run AI models quickly and at low power.
Prefill throughput
Prefill throughput is how fast an AI model reads its input before it starts writing, measured in tokens per second.
