What Is a Vector? The Shape of an Embedding
DICTIONARY · AI

What Is a Vector in AI?

A vector is an ordered list of numbers, and in AI it is the form content takes once it has been embedded.

In plain English

An embedding is a vector. The word describes the shape of the data: a fixed-length list of numbers, often hundreds or thousands of them.

Each position contributes to locating the content in a space where distance corresponds to similarity. That is the whole reason the representation is useful.

What to know

A list of numbers
Fixed length, defined by the model that produced it.
Dimensions
Each number is one dimension of the representation.
Distance means similarity
Cosine similarity is the usual measure.
Comparable within a model
Vectors from different models occupy different spaces.

Why it matters

The term appears constantly in AI architecture discussions, and knowing that it simply means the numeric form of a piece of content removes most of the mystery from vector databases and semantic search.

Common mistakes

×Comparing vectors produced by different models.
×Assuming more dimensions is automatically better.
×Treating vector similarity as a factual guarantee.
×Confusing it with a vector graphic, which is unrelated.

FAQs

Is a vector the same as an embedding?

An embedding is a vector produced to represent meaning. The terms overlap in practice.

Why cosine similarity?

It compares direction rather than magnitude, which suits meaning comparison.

WRITTEN BY TARIQ SALLAM
Marketing Consultant. Entrepreneur. Content Creator.

I'm a marketing consultant, entrepreneur and content creator. I help businesses grow through practical marketing, websites, SEO, content and AI.

More About Tariq →