Name: retrieval-rerank
Price: 0.005 USDC
Availability: InStock

$ man retrieval-rerank

agentutility / wordmint / retrieval-rerank

PRICE / CALL

$0.005

USDC · base mainnet · scheme: exact

METHOD

POST

CLUSTER

wordmint

CATEGORY

uncategorized

STATUS

● live

NAME

retrieval-rerank — reranks retrieval results: given a query and up to 30 candidate documents, scores each 0-100 for query relevance using venice qwen3-5-35b…

SYNOPSIS

POST https://x402.agentutility.ai/retrieval-rerank
     Content-Type: application/json
     X-PAYMENT:    <signed-transferWithAuthorization>

     { ... }

↳ first call → 402 Payment Required. Sign USDCtransferWithAuthorization, retry with theX-PAYMENT header.

DESCRIPTION

Reranks retrieval results: given a query and up to 30 candidate documents, scores each 0-100 for query relevance using Venice qwen3-5-35b-a3b in JSON-mode and returns sorted top-k. Useful as a second-stage reranker on top of cheap vector retrieval. Use it as a RAG reranker, document scoring endpoint, top-k filter, or cross-encoder substitute.

INPUT — request schema

property	type	description	req?
query	string	Search query.	required
documents	array	Candidate documents. Up to 30; each up to 2,000 chars (truncated).	required
top_k	number	How many top documents to return. Default min(5, len(documents)).	optional

OUTPUT — response shape

field	type	description
query	string	Echo of the original query string the documents were scored against.
ranked	string	Array of {index, score, document} objects sorted by relevance score descending, length up to top_k.
doc_count	string	Number of candidate documents submitted in the request before reranking.
top_k	string	Number of top-scored documents returned in the ranked array.
model	string	Venice model ID used for scoring, typically qwen3-5-35b-a3b in JSON mode.
source	string	Upstream provider tag identifying where the reranking inference ran, e.g. 'venice'.

EXAMPLES — two ways to call

EXAMPLE 1 · curl

curl -X POST https://x402.agentutility.ai/retrieval-rerank \
  -H 'Content-Type: application/json' \
  -d '{ }'

first response = 402 Payment Required with payment requirements; sign + retry with X-PAYMENT.

EXAMPLE 2 · mcp

# Install the MCP package for this endpoint's cluster
npx -y @agentutility/mcp-<cluster>

# Required: EVM private key with USDC on Base
export X402_PRIVATE_KEY=0x...

# Then call the retrieval-rerank tool from your MCP-aware agent.

MCP server handles payment automatically — your coding agent just calls the tool by name.

METADATA

tags: wordmintragrerankerretrievaldocument-scoringtop-kcross-encoder
methods: POST
cluster: wordmint
price: $0.005 USDC per call

ADJACENT — other endpoints in wordmint

endpoint	description	price
brand-tagline	Generates brand taglines and slogans for launch pages, X bios, email copy, and product cards.	$0.005
brand-tagline-generate	Generates tagline options for a brand or startup from its name, concept, audience, and tone.	$0.005
card-resolve	Normalizes free-form graded card text into a canonical card object.	$0.005
content-simhash	Fingerprints text with a 64-bit SimHash for near-duplicate detection, computed entirely locally.	$0.005
cron-parse	Cron parser.	$0.005
detect-language	Language detector / language identification.	$0.005
dictionary-define	Looks up English word definitions with pronunciation, part of speech, and synonyms.	$0.005
embedding-similarity	Measures how semantically similar two strings are: embeds both via Venice (default model: text-embedding-bge-m3) and returns the cosine s…	$0.005