Generate Local Embeddings

What are embeddings?

Embeddings turn text into lists of numbers. Similar text gets similar numbers. They power search, recommendations, and retrieval-augmented generation (RAG).

Pull an embedding model

ollama pull nomic-embed-text

Generate an embedding

curl http://localhost:11434/api/embeddings -d '{
  "model": "nomic-embed-text",
  "prompt": "How do I reset my password?"
}'

Build a tiny RAG pipeline in Python

import requests, numpy as np

def embed(text):
    r = requests.post("http://localhost:11434/api/embeddings", json={
        "model": "nomic-embed-text",
        "prompt": text
    })
    return np.array(r.json()["embedding"])

def similarity(a, b):
    return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))

docs = [
    "Reset your password on the settings page.",
    "Our API supports OAuth2 authentication.",
    "Deploy with docker compose up."
]

vectors = [embed(d) for d in docs]
query = embed("How do I change my password?")
scores = [similarity(query, v) for v in vectors]
best = docs[np.argmax(scores)]
print(best)

Why local embeddings matter

Private documents stay on your machine
No per-token embedding charges
Fast for small-to-medium document sets

For larger collections, move vectors to a local vector database like Chroma or pgvector.