AI search API

Real-time web search, distilled for LLMs.

ODEN turns any query into clean, citation-backed JSON your model can read. One single API call delivers grounded synthesis — no HTML noise, no scraped clutter, and webmaster opt-out honored at the source.

POST /search
$ curl https://api.oden-api.com/search \
    -H "Authorization: Bearer $ODEN_KEY" \
    -d '{ "query": "What are the symptoms of vitamin D deficiency?" }'
{
  "query": "What are the symptoms of vitamin D deficiency?",
  "results": {
    "answer": "Low vitamin D can cause muscle weakness, bone pain and fatigue, and in severe cases rickets in children or osteomalacia in adults. Mild deficiency is often symptom-free and found only through a blood test.",
    "citations": [
      { "title": "Vitamin D deficiency — 1177 Vårdguiden",
        "url":   "https://www.1177.se/...",
        "score": 0.91 }
    ]
  }
}

Every response pairs a concise synthesis with ranked source citations, including destination URLs, page titles, and relevance scores that your application can inspect before using the answer.

Traceable sources

Answers are grounded in web pages retrieved for the query. ODEN returns destination URLs, titles, relevance scores, and available publication metadata so your application can show its sources and let users verify important claims.

Source-aware retrieval

ODEN checks robots.txt and supported machine-readable TDM reservation signals before fetching a page. Citations are returned as metadata with links, helping teams build a more reviewable data-sourcing process.

Distilled for Context

Receive concise, high-density synthesis alongside ranked citation objects. Clean JSON structures eliminate HTML tags, boilerplate script payloads, and navigation noise, minimizing prompt token consumption.

Edge Latency
Edge-routed requests

Requests enter through Cloudflare's edge network. Cached lookups can return quickly, while live retrieval depends on the response time of external source sites.

Data Privacy
No advertising profiles

ODEN does not sell search queries to data brokers or use customer requests to train a public foundation model. Operational handling is described in the privacy policy.

Tool Calling
Native Schema Support

Built-in JSON schemas ready for immediate drop-in integration with OpenAI function calling, Claude tools, and LangChain.

Swedish Edge
Built in Sweden

Engineered and operated from Söderhamn, Sweden, with privacy documentation and source-handling controls designed for European teams.

Pricing

Start free. Scale when your agents do.

Free
€0

1,000 searches / month

  • Synthesized answer + citations
  • Full retrieval pipeline
  • robots.txt & TDM opt-out checks
  • No card required
Get a key
Pro
€29 / month

6,000 searches / month

  • Everything in Free
  • 6× the monthly quota
  • Full retrieval & extraction pipeline
  • Cancel anytime without lock-in
Upgrade in dashboard
Top-up
€7 / 1,000

Pay-as-you-go · never expires

  • Add searches on demand
  • Stacks on any plan
  • Spent only after monthly quota
  • One-time, no subscription
Buy top-ups

All accounts start with 1,000 free searches refreshed every calendar month with zero credit card required. Upgrade to Pro for high-volume agent workflows, or buy non-expiring top-ups that carry over indefinitely. Quota is deducted strictly on successful queries.

Also from ODEN

Same retrieval engine — no integration required.

ODEN AI is the chat interface: ask a question, get a synthesized answer with citations, powered by the same search API. Free tier included.

Interactive Research

Ask exploratory questions and receive web-grounded answers with source links rendered inline for review.

Streaming Synthesis

Tokens stream instantly via Server-Sent Events (SSE) with live progress indicators as sources are fetched.

Zero Setup

No SDKs, API keys, or infrastructure to configure. Sign in with Google or GitHub and start exploring immediately.

Odin kept two ravens — Huginn and Muninn, thought and memory — who flew across the world each day and returned with everything they had seen. ODEN does the same for you.

Engineered for software engineers, ML practitioners, and AI agents that need compact web retrieval with traceable sources. ODEN combines a documented retrieval pipeline, transparent usage tiers, and links users can inspect before acting on an answer.

Go deeper: read the API documentation, browse integration guides for RAG and tool use, see how ODEN compares to Tavily, SerpAPI and others, or check the pricing.