# JarvisClaw Rerank

> JarvisClaw Rerank is a paid API for AI agents from api.jarvisclaw.ai, paid per call via x402, $0.045/call, status unknown (last checked 2026-09-13).

Reranks a list of documents or passages by relevance to a query using AI, via a pay-per-call x402 gateway settled in USDC

## Facts

- Endpoint: POST https://api.jarvisclaw.ai/v1/rerank
- Price: $0.045/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-13
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/jarvisclaw-rerank-8b336754
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_R-OHdCco4K79i2xl1QNE7

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 jarvisclaw-rerank-8b336754 -d '<json body>'
```

Example prompt: Rerank these 10 retrieved passages by how relevant they are to the query 'What are the side effects of ibuprofen?' and return them sorted with relevance scores so I can pass the top 3 to my LLM.

## When to prefer this

Use this endpoint when you have a set of candidate documents or passages retrieved via vector search, keyword search, or any retrieval method, and want to rerank them by semantic relevance before feeding them into an LLM. Ideal for RAG pipelines where retrieval quality matters. Prefer this over embedding-based similarity when you need a dedicated cross-encoder reranker. The x402 pay-per-call model suits sporadic or agent-driven usage without subscription commitment.

## Known failure modes

- Empty document list returns an error or empty results
- Query or documents exceed token/length limits causing a 400 error
- Payment not included or insufficient USDC balance triggers a 402 Payment Required response
- Malformed request body (missing query or documents field) returns a 422 validation error
- Service unavailable or gateway timeout returns a 5xx error
- Documents list too large causes rate limit or payload error

## How this service works

AI API Gateway with smart routing, pay per call via x402. OpenAI-compatible. Settled in USDC on Base & Solana.

## Output

A reranked list of the input documents sorted by relevance score descending, each item accompanied by a relevance/confidence score indicating how well it matches the query.

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/jarvisclaw-rerank-8b336754/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.jarvisclaw.ai](https://www.zero.xyz/host/api.jarvisclaw.ai/llms.txt)