Embedding Model

ELI5 — The Vibe Check

An embedding model turns words, sentences, or documents into lists of numbers (vectors) that capture meaning. Similar things end up close together in this number space — 'king' is near 'queen,' 'Python' is near 'programming.' It's like giving every concept a GPS coordinate in meaning-space. This is how AI search, recommendations, and RAG actually work under the hood.

Real Talk

An embedding model maps text (or other data) to dense vector representations in a high-dimensional space where semantic similarity correlates with vector proximity. Modern text embedding models (like OpenAI's text-embedding-3, Cohere's Embed, BGE) produce vectors of 256-3072 dimensions. They're used for semantic search, clustering, classification, and as the retrieval component in RAG systems.

Show Me The Code

# Generate embeddings with OpenAI
from openai import OpenAI
client = OpenAI()

response = client.embeddings.create(
    model="text-embedding-3-small",
    input="What is Docker?"
)
vector = response.data[0].embedding  # 1536 floats

When You'll Hear This

"We use an embedding model to power semantic search." / "The embedding space clusters similar support tickets together."