AI Wiki

A2A

Agent2Agent

A2A stands for Agent2Agent. In an AI context, it usually refers to a class of protocols or mechanisms that let different agents discover, communicate with, collaborate with, and hand off tasks to one another.

Tool Calling

Tool Calling is also often called Function Calling. It's the mechanism that lets a large language model request help from external tools, and it's an important foundation for an Agent to execute tasks.

Local Model

A local model is an AI model that runs on your own computer or server, so requests don't have to go to a cloud provider. In 2026 it moved from "can I get it running" to "splitting work with the cloud": sensitive steps stay local, hard problems go to the cloud.

MCP

Model Context Protocol

MCP stands for Model Context Protocol. It's an open protocol that lets AI applications connect to external tools and data sources, and since late 2025 it has been hosted by the Agentic AI Foundation under the Linux Foundation.

VPN

Virtual Private Network

VPN stands for Virtual Private Network. It isn't an AI-specific term, but you'll often run into it when using AI tools, APIs, relays, local services, and enterprise systems.

API Relay

An API relay is a term many AI tool users have run into recently. It usually refers to a transfer platform that sits between the user and the model's official service: instead of connecting directly to the official interfaces of OpenAI, Anthropic, Google, and others, the user calls the models through a unified address and key provided by the relay.

API Key

An API key is a string of characters you'll often see when using AI interfaces, plugins, and relays. It looks like a random password, but it's essentially a calling credential the service provider issues to you.

Agent

In an AI context, an agent is a system that doesn't just answer questions but plans steps toward a goal, calls tools, observes results, and keeps acting. In 2026 agents moved from demos into everyday work, and the hard part shifted from "can it do this" to "how do we run it reliably and keep it in check."

Model Parameters

Model parameters are an unavoidable term when understanding "model size." We often hear that a model has billions, tens of billions, or hundreds of billions of parameters; these parameters aren't rules written in by users, but a large set of numeric values the model learns during training.

API

Application Programming Interface

API stands for Application Programming Interface. It's a set of conventions by which software programs request services from one another and exchange data.

Multimodal AI

In the past, many AI tools handled only text, or only images. What sets multimodal AI apart is that it can understand and generate multiple forms of information at once, such as text, images, audio, and video.

AI Hallucination

In an AI context, hallucination refers to a model generating content that looks fluent, confident, and real, but that is actually wrong, fabricated, unsupported, or inconsistent with the given materials.