Embeddings - Planck

POST /v1/embeddings converts one string or a batch of strings into vectors. Use it for semantic search, retrieval, clustering, recommendations, and similarity comparisons.

Create an Embedding

TypeScript
Python
cURL

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inquantum.ai/v1",
  apiKey: process.env.PLANCK_API_KEY,
});

const response = await client.embeddings.create({
  model: "text-embedding-3-small",
  input: ["first document", "second document"],
});

console.log(response.data[0].embedding);

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.inquantum.ai/v1",
    api_key=os.environ["PLANCK_API_KEY"],
)

response = client.embeddings.create(
    model="text-embedding-3-small",
    input=["first document", "second document"],
)

print(response.data[0].embedding)

curl https://api.inquantum.ai/v1/embeddings \
  -H "Authorization: Bearer $PLANCK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-3-small",
    "input": ["first document", "second document"]
  }'

The response follows the OpenAI embeddings shape with an ordered data array and input-token usage.

Choosing a Model

Use a model that advertises the embeddings operation and copy its model ID from the model registry. The Planck endpoint, authentication, and response shape stay the same regardless of the selected model.

Anthropic does not expose an embeddings API. The Anthropic-compatible Messages route is for text generation, not embeddings.

Billing and Tracking

Embedding requests are recorded in Requests with the selected provider, model, input tokens, duration, cost, request body, and response body. If you send the optional user field, it is used as the request’s user attribution. See the Embeddings API reference for request and response fields.