On this page

• Basic request
• Embedding multiple texts at once
• Comparing similarity
• Choosing a model

Embeddings Guide

Embeddings turn text into a vector of numbers that capture its meaning — useful for semantic search, clustering, recommendations, and RAG (retrieval-augmented generation). Available models: text-embedding-3-large, text-embedding-3-small, text-embedding-ada-002.

POST https://api.oblion.io/v1/embeddings

Basic request

curl https://api.oblion.io/v1/embeddings \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "text-embedding-3-small",
    "input": "The food was delicious and the service was excellent."
  }'

from openai import OpenAI

client = OpenAI(base_url="https://api.oblion.io/v1", api_key="YOUR_API_KEY")

response = client.embeddings.create(
    model="text-embedding-3-small",
    input="The food was delicious and the service was excellent.",
)
vector = response.data[0].embedding
print(len(vector), vector[:5])

import OpenAI from "openai";

const openai = new OpenAI({ baseURL: "https://api.oblion.io/v1", apiKey: "YOUR_API_KEY" });

const response = await openai.embeddings.create({
  model: "text-embedding-3-small",
  input: "The food was delicious and the service was excellent.",
});
console.log(response.data[0].embedding.length);

Embedding multiple texts at once

Pass an array to input to batch multiple strings in a single request — the response data array preserves the same order via each item's index:

response = client.embeddings.create(
    model="text-embedding-3-small",
    input=["cat", "dog", "airplane"],
)
for item in response.data:
    print(item.index, item.embedding[:3])

Comparing similarity

Use cosine similarity to measure how close two embeddings are — closer to 1 means more similar:

import numpy as np

def cosine_similarity(a, b):
    a, b = np.array(a), np.array(b)
    return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))

response = client.embeddings.create(
    model="text-embedding-3-small",
    input=["The cat sat on the mat.", "A feline rested on the rug.", "Quarterly revenue exceeded expectations."],
)
vectors = [d.embedding for d in response.data]

print(cosine_similarity(vectors[0], vectors[1]))  # similar sentences -> higher score
print(cosine_similarity(vectors[0], vectors[2]))  # unrelated sentence -> lower score

Choosing a model