Embeddings

The Embedding API converts text into vector representations that capture semantic meaning. Use embeddings for semantic search, clustering, and RAG applications.

What you’ll learn:

How to generate embeddings from text
Available embedding models and their use cases
How to use embeddings for semantic search and RAG

Basic Usage

curl -X POST "$OPENAI_BASE_URL/embeddings" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "jina-embeddings-v2-base-de",
    "input": ["The quick brown fox jumps over the lazy dog", "Data science is fun!"]
  }'

from openai import OpenAI

client = OpenAI()

texts = [
    "I am Batman and I'm rich",
    "I am Spiderman",
    "I am Ironman and I'm a billionaire",
    "I am Flash",
    "I am the president of USA",
]

embeddings = client.embeddings.create(input=texts, model="jina-embeddings-v2-base-de")

print("Embedding dimension:", len(embeddings.data[0].embedding))
print("Number of vectors:", len(embeddings.data))
print("Token usage:", embeddings.usage)

import OpenAI from "openai";

const client = new OpenAI();

const result = await client.embeddings.create({
  model: "jina-embeddings-v2-base-de",
  input: ["The quick brown fox jumps over the lazy dog", "Data science is fun!"],
});

console.log("Embedding dimension:", result.data[0].embedding.length);
console.log("Token usage:", result.usage);

Example output:

Embedding dimension: 768
Number of vectors: 5
Token usage: Usage(prompt_tokens=41, total_tokens=41)

Available Embedding Models

Model	Dimensions	Languages	Use Case
`jina-embeddings-v2-base-de`	768	German & English	German-language RAG
`text-embedding-bge-m3`	1024	Multilingual	Cross-language search

Check Available Models for the latest list.

Next Steps

LangChain Integration — Use embeddings with LangChain for RAG
LlamaIndex Integration — Use embeddings with LlamaIndex for RAG