# PennyRail Embedding API (ai.v1-embeddings)

> PennyRail Embedding API (ai.v1-embeddings) is a paid API for AI agents from pennyrail.vercel.app, paid per call via x402, $0.002/call, status unknown (last checked 2026-11-01).

Generates vector embeddings for input data via a paid micropayment-gated API endpoint on PennyRail

## Facts

- Endpoint: POST https://pennyrail.vercel.app/api/p/mini/ai.v1-embeddings--embedding-api?utm_source=zero.xyz
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-11-01
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/pennyrail-embedding-api-ai-v1-embeddings-8a8e835d
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_9-A3jQy9TsoSjHqkYrDKh

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 pennyrail-embedding-api-ai-v1-embeddings-8a8e835d -d '<json body>'
```

Example prompt: Can you generate a vector embedding for this text: 'The quick brown fox jumps over the lazy dog' — I need it for a semantic similarity pipeline I'm building?

## When to prefer this

Choose this endpoint when you need a micropayment-gated, x402-compatible embedding API that integrates natively with Coinbase's payment infrastructure. It is well-suited for AI agents that already operate within the x402 pay-per-call ecosystem and need to generate embeddings on demand without a pre-provisioned API key or subscription. Prefer alternatives like OpenAI or Cohere embeddings if you need guaranteed SLAs, well-documented schemas, or high-volume batch processing.

## Known failure modes

- Payment not provided or insufficient USDC (402 Payment Required)
- Malformed or missing 'input' field in request body (400 Bad Request)
- Input object type not supported by the underlying embedding model
- Service unavailable or cold-start latency on Vercel serverless (503/504)
- Response schema is fully open-ended, so embedding field name must be inferred at runtime

## How this service works

Machine-readable settlement service

## Output

Returns an object containing the computed embedding vector (numeric float array) representing the semantic content of the input, usable directly in vector databases, similarity computations, or ML pipelines. The response schema is open-ended, so additional metadata fields may be included alongside the embedding.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "additionalProperties": true
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/pennyrail-embedding-api-ai-v1-embeddings-8a8e835d/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from pennyrail.vercel.app](https://www.zero.xyz/host/pennyrail.vercel.app/llms.txt)