# SiteCheck Rerank

> SiteCheck Rerank is a paid API for AI agents from api.sitecheck-api.workers.dev, paid per call via x402, $0.002/call, status unknown (last checked 2026-10-19).

Reranks a list of documents by relevance to a query using a neural reranker model, returning ranked results with relevance scores

## Facts

- Endpoint: POST https://api.sitecheck-api.workers.dev/api/rerank?utm_source=zero.xyz
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-19
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/sitecheck-rerank-8bf22649
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_izsb7NuzWBwKu-fax-yXS

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 sitecheck-rerank-8bf22649 -d '<json body>'
```

Example prompt: Rerank these 5 document passages by how well they answer the question 'how does x402 payment work?' and return them sorted by relevance score.

## When to prefer this

Use this endpoint when you need a lightweight, pay-per-call cross-encoder reranker to improve retrieval precision in RAG pipelines or hybrid search systems without setting up your own reranking infrastructure. It is ideal for agents that retrieve candidate documents via vector search and need to reorder them by true semantic relevance before passing to an LLM. Prefer this over embedding-based similarity when you need a dedicated cross-encoder that jointly encodes query and document for higher-quality relevance scores.

## Known failure modes

- More than 50 documents submitted — returns error
- Document exceeds 2,000 character limit — may be truncated or rejected
- Query exceeds 500 character limit — returns validation error
- Empty documents array — returns error
- Payment not included or insufficient — 402 Payment Required
- Model inference timeout on large batches — may return 500 error

## How this service works

Orders up to 50 documents by relevance to a query with the BGE reranker and returns rank, index and score for each. Use it after a keyword or vector search to pick the best few. Scores compare documents within one call only.

## Output

Returns a JSON object containing the reranked documents sorted by relevance, with each result including its rank position (1-based), original index in the input array, a floating-point relevance score, and the document text. Also includes the total count, the model used (bge-reranker-base), and a timestamp.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "query": {
   "type": "string",
   "description": "The question or search, up to 500 characters"
  },
  "documents": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "description": "1 to 50 documents, up to 2,000 characters each"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "count": 2,
  "model": "@cf/baai/bge-reranker-base",
  "query": "how do I pay per call",
  "results": [
   {
    "rank": 1,
    "index": 1,
    "score": 0.057772,
    "document": "x402 lets an agent pay per request in USDC, with no account."
   },
   {
    "rank": 2,
    "index": 0,
    "score": 0.000037,
    "document": "Our office is closed on public holidays."
   }
  ],
  "checkedAt": "2026-10-14T05:12:26.366Z"
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/sitecheck-rerank-8bf22649/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.sitecheck-api.workers.dev](https://www.zero.xyz/host/api.sitecheck-api.workers.dev/llms.txt)