Exa Search API

Exa (formerly Metaphor) is a semantic search engine that uses a large-scale neural network to understand context. It is designed to find high-quality content that traditional keyword search engines might miss.

Endpoint

POST https://llm.siraya.pro/v1/exa

Request Parameters

Parameter	Type	Required	Description
model	string	Yes	Use exa-search.
query	string	Yes	The search query.
type	string	No	keyword, neural, magic, or auto.
category	string	No	Filter by category (e.g., company, news, github, pdf).
numResults	integer	No	Number of results to return (default: 10).
includeText	boolean	No	Whether to include the full text of the pages.

Example Request

cURLPython

curl https://llm.siraya.pro/v1/exa \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <API_KEY>" \
  -d '{
    "model": "exa-search",
    "query": "The future of quantum computing",
    "type": "neural",
    "numResults": 5
  }'

import requests

url = "https://llm.siraya.pro/v1/exa"
headers = {"Authorization": "Bearer <API_KEY>"}
data = {
    "model": "exa-search",
    "query": "The future of quantum computing",
    "numResults": 5
}

response = requests.post(url, headers=headers, json=data)
print(response.json())

Response

Returns a JSON object containing the requestId, confirmed search type, and an array of results with document titles, URLs, and metadata.