# OpenAI Embeddings

> OpenAI Embeddings is a paid API for AI agents from OpenAI at openai.mpp.tempo.xyz, paid per call via MPP, $0.0001/call, status unknown (last checked 2026-10-11, last successful call 2026-05-28).

Converts text into high-dimensional vector embeddings for semantic search, clustering, and similarity tasks

## Facts

- Endpoint: POST https://openai.mpp.tempo.xyz/v1/embeddings?utm_source=zero.xyz
- Price: $0.0001/call
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-11
- Last successful call: 2026-05-28
- Success rate: 100% of calls made through Zero
- Activations on Zero: 2
- Provider: OpenAI
- Docs: https://platform.openai.com/docs
- Website: https://openai.com
- Tags: llm, gpt-4o, dall-e, whisper, tts, embeddings, chat, ai, media
- Canonical page: https://www.zero.xyz/c/openai-embeddings-21fb8481
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_NmfbyIYsrPUxIRZVx2bH1

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability openai-embeddings-21fb8481 -d '<json body>'
```

Example prompt: Generate an embedding vector for the text 'The quick brown fox jumps over the lazy dog' using the text-embedding-ada-002 model so I can store it in my vector database for semantic search.

## When to prefer this

Choose this endpoint when you need to convert text into numerical vector representations for downstream tasks like semantic search, clustering, classification, or storing in a vector database. Ideal for RAG (Retrieval-Augmented Generation) pipelines, recommendation systems, and any workflow requiring semantic similarity computation. Prefer over alternatives when you need OpenAI's text-embedding models (e.g. text-embedding-ada-002, text-embedding-3-small, text-embedding-3-large) specifically.

## Known failure modes

- Invalid or missing input text returns a 400 error
- Exceeding token limit for the chosen model returns a context-length error
- Invalid model name returns a 404 or model-not-found error
- Rate limiting returns a 429 too-many-requests error
- Empty input array returns a validation error
- Input contains unsupported data type (non-string) causes a 400 bad request

## How this service works

Create embeddings

## Output

Returns a JSON object containing a list of embedding objects, each with a dense float vector representing the input text, along with metadata including the model used and token usage statistics (prompt tokens and total tokens consumed).

## Example request

```json
{
 "input": "The quick brown fox jumps over the lazy dog",
 "model": "text-embedding-3-small",
 "voice": "alloy",
 "response_format": "json"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/openai-embeddings-21fb8481/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from openai.mpp.tempo.xyz](https://www.zero.xyz/host/openai.mpp.tempo.xyz/llms.txt)