API v1 · Open Beta

HyperSearchX Documentation

HyperSearchX is an intelligent search API that aggregates multiple search backends, extracts structured content from web pages, runs multi-step research pipelines, and delivers token-budgeted, AI-ready context through a clean REST API.

Built in Rust for maximum performance, it implements 17 novel algorithms including HyperFusion (8-signal neural ranking), CEP (5-layer content extraction), and QATBE (query-aware token budget extraction).

Open Beta: HyperSearchX is currently in open beta. The Free tier provides 1,000 requests/month with no credit card required. Get your API key →

Quick links

⚡

Quick Start

Make your first API call in under 2 minutes.

🔑

Authentication

Bearer tokens, admin secrets, key rotation.

🔍

Search API

Multi-backend federated search with neural ranking.

🔬

Research API

Deep multi-source research with citations.

⏱

Rate Limits

Plan quotas, per-minute limits, headers.

🐳

Self-Hosting

Run on your own infrastructure for free.

Base URL

All API requests are made to the following base URL:

text

https://api.hypersearchx.zuhabul.com

Your first request

Once you have an API key from the dashboard, you can make your first search request:

search.sh

curl -X POST https://api.hypersearchx.zuhabul.com/v1/search \
  -H "Authorization: Bearer hsx_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "rust async programming best practices",
    "tier": "detailed",
    "max_sources": 10
  }'

response.json

{
  "meta": {
    "query": "rust async programming best practices",
    "tier": "detailed",
    "tokens_used": 2847,
    "sources_count": 10,
    "duration_ms": 1243,
    "result_id": "09742460-58db-44fd-9036-502ed26565e2"
  },
  "results": [
    {
      "title": "Async in Rust: A comprehensive guide",
      "url": "https://tokio.rs/tokio/tutorial",
      "snippet": "Tokio is an asynchronous runtime for the Rust programming...",
      "score": 0.923
    }
  ]
}

Key concepts

Detail tiers

Every search and research request accepts a tier parameter that controls how much content is extracted and how many tokens are returned:

Tier	Tokens (approx)	Use case
`key_facts`	~200	Quick answers, chatbot responses
`summary`	~1,000	General AI context, RAG pipelines
`detailed`	~5,000	Thorough analysis, long-form content
`complete`	~20,000	Full document extraction

Token budget

You can override the tier with an explicit token_budget (100–10,000 for search, 1,000–50,000 for research). The API uses the QATBE algorithm to maximize relevant content within your budget.

Authentication

All API endpoints (except /health) require a Bearer token in the Authorization header. API keys are prefixed with hsx_ and contain 64 hex characters (256-bit entropy).

Endpoints overview

Method	Endpoint	Description
`POST`	`/v1/search`	Multi-backend web search with neural ranking
`POST`	`/v1/scrape`	URL content extraction with CEP pipeline
`POST`	`/v1/research`	Full multi-source research with citations
`POST`	`/v1/youtube/search`	YouTube video search & analysis
`POST`	`/v1/social/research`	Cross-platform social media research
`GET`	`/v1/usage`	Your API usage statistics & quota
`GET`	`/health`	Service health & dependency status

Rate limits & quotas

Plan	Requests / month	Requests / minute	Price
Free	1,000	60	$0
Starter	10,000	200	$19 / mo
Pro	100,000	500	$79 / mo
Enterprise	Unlimited	2,000	Custom

When rate-limited, the API returns 429 Too Many Requests with a Retry-After: 60 header. See Rate Limits for full details.

SDKs

Official SDKs are coming soon. In the meantime, use the REST API directly or with any HTTP client.

hypersearchx.ts

// TypeScript — plain fetch
const HSX_BASE = "https://api.hypersearchx.zuhabul.com";

async function search(query: string, tier = "summary") {
  const res = await fetch(`${HSX_BASE}/v1/search`, {
    method: "POST",
    headers: {
      "Authorization": `Bearer ${process.env.HSX_API_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({ query, tier, max_sources: 10 }),
  });
  if (!res.ok) throw new Error(`HSX API error: ${res.status}`);
  return res.json();
}

const result = await search("best vector databases 2025", "detailed");
console.log(result.results[0].title);

hypersearchx.py

# Python — requests
import os, requests

HSX_BASE = "https://api.hypersearchx.zuhabul.com"
HEADERS = {"Authorization": f"Bearer {os.environ['HSX_API_KEY']}"}

def search(query: str, tier: str = "summary") -> dict:
    r = requests.post(f"{HSX_BASE}/v1/search",
        headers=HEADERS,
        json={"query": query, "tier": tier, "max_sources": 10})
    r.raise_for_status()
    return r.json()

result = search("best vector databases 2025", "detailed")
print(result["results"][0]["title"])

🚀

Ready to start building?

Get your free API key and make your first request in under 2 minutes.

Get API Key →Quick Start Guide