POST /v1/embeddings: Embeddings

Embeddings in the OpenAI shape, through your own OpenAI or Google key (or an OpenAI-compatible endpoint you configured), never the platform's. Memory itself needs no key of yours: this endpoint is for your own use of embeddings.

Example

Against https://platform.memtro.com, with your key in MEMTRO_API_KEY.

curl https://platform.memtro.com/v1/embeddings \ -H "Authorization: Bearer $MEMTRO_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model": "openai/text-embedding-3-small", "input": ["billing migration", "Q3 launch"]}'

from openai import OpenAI client = OpenAI(base_url="https://platform.memtro.com/v1", api_key="sk-memtro-…") result = client.embeddings.create(model="openai/text-embedding-3-small", input=["billing migration", "Q3 launch"]) print(len(result.data[0].embedding))

import OpenAI from "openai"; const client = new OpenAI({ baseURL: "https://platform.memtro.com/v1", apiKey: process.env.MEMTRO_API_KEY }); const result = await client.embeddings.create({ model: "openai/text-embedding-3-small", input: ["billing migration", "Q3 launch"] }); console.log(result.data[0].embedding.length);

Vectors shortened here.

{
  "object": "list",
  "data": [
    { "object": "embedding", "index": 0, "embedding": [0.0123, -0.0481, …] },
    { "object": "embedding", "index": 1, "embedding": [-0.0067, 0.0215, …] }
  ],
  "model": "openai/text-embedding-3-small",
  "usage": { "prompt_tokens": 6, "total_tokens": 6 }
}

Authentication

An API key, as Authorization: Bearer or x-api-key.

Every reply, errors included, carries x-request-id and request-id headers with the same id, starting req-. Quote it to support.
It sends no CORS headers, so a page on another origin cannot call it; call it from a server.

Request body

JSON, with Content-Type: application/json.

modelstringOptional

An embedding model id. memtro and brain mean the default.

Default:

The server's embedding model, then openai/text-embedding-3-small, then google/gemini-embedding-001, whichever you hold a key for

inputstring or array of stringsRequired

The text to embed, or several.

encoding_formatstringOptional

Accepted, but the reply is always arrays of floats.

One of float or base64.

Response

200application/json

One vector per input, in order.

object"list"Optional: Always list.
dataarray of objectsOptional: {"object": "embedding", "index", "embedding": [numbers]} per input. OpenAI and Google models are asked for vectors of the server's configured dimension.
modelstringOptional: <provider>/<model> that made them.
usageobjectOptional: prompt_tokens, total_tokens.

Errors

In the OpenAI envelope, {"error": {"message", "type", "param", "code"}}. What each code means and what to do about it is in the error catalogue.

Status	Code	When
400	none	The body is not `{ input: string \| string[] }`.
400	`provider_not_configured`	No embedding-capable key: add an OpenAI or Google key.
400	`invalid_provider_base_url`	An OpenAI-compatible endpoint's base URL cannot be used: not a URL, not `https://`, a user name or password in it, or a port not accepted.
400	`provider_base_url_not_allowed`	An OpenAI-compatible endpoint's host is not approved for the organisation, is private or internal, does not resolve, or redirected, whether refused before the request or during it.
401	`invalid_api_key`	The key is missing, malformed, revoked or expired.
402	`subscription_inactive`	The organisation has no active subscription.
provider's	`upstream_error`	The provider returned an error.

Notes

No CORS headers: call it from a server, not from a page on another origin.

Embeddings