# Text Embedding API (768-dim)

> Text Embedding API (768-dim) is a paid API for AI agents from river-guestbook-newark-fails.trycloudflare.com, paid per call via x402, $0.005/call, status unknown (last checked 2026-10-16).

Converts text input (single string or batch) into 768-dimensional vector embeddings for semantic tasks

## Facts

- Endpoint: POST https://river-guestbook-newark-fails.trycloudflare.com/v1/embed?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-16
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/text-embedding-api-768-dim-1a54c097
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_89mpUNlVqvqHqvMXYb0ZR

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 text-embedding-api-768-dim-1a54c097 -d '<json body>'
```

Example prompt: Turn this sentence into a 768-dimensional embedding vector so I can compare it semantically against my document database: 'The quick brown fox jumps over the lazy dog.'

## When to prefer this

Choose this endpoint when you need lightweight, fast 768-dimensional text embeddings for semantic search, clustering, or similarity tasks, especially in agentic pipelines that support x402 micropayment authentication. The 768-dim output is compatible with many standard vector stores (Pinecone, Weaviate, Qdrant). Avoid if you need higher-dimensional embeddings (e.g. 1536-dim OpenAI style) or if you require a guaranteed SLA from a named commercial provider.

## Known failure modes

- Empty or missing 'input' field returns an error
- Input text exceeding token limit may be truncated or rejected
- Malformed JSON body returns 400-level error
- Temporary Cloudflare tunnel unavailability returns 5xx or connection timeout
- Payment failure or insufficient USDC balance blocks the request

## How this service works

Text embeddings (768-dim). Send {"input": "text"} or {"input": ["a","b"]}.

## Output

Returns a JSON object containing a 'choices' array with the embedding vector(s) (768-dimensional float arrays) for each input text, along with a 'usage' object reporting token counts consumed.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "usage": {
   "type": "object"
  },
  "choices": {
   "type": "array"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/text-embedding-api-768-dim-1a54c097/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from river-guestbook-newark-fails.trycloudflare.com](https://www.zero.xyz/host/river-guestbook-newark-fails.trycloudflare.com/llms.txt)