cd /news/ai-tools/show-hn-any-search-one-api · home topics ai-tools article
[ARTICLE · art-31788] src=searchrouter-demo.vercel.app ↗ pub= topic=ai-tools verified=true sentiment=↑ positive

Show HN: Any Search, One API

SearchRouter launches a unified API that lets developers switch between search providers like Exa, Cohere, and Brave with a single string change, offering automatic fallback, fan-out routing, and unified billing. The platform aims to simplify the fragmented search landscape for AI agents and applications.

read2 min views3 publishedJun 17, 2026

One API.

Search is evolving, and every agent needs different search. Developers need choice. SearchRouter puts every provider behind one API, with automatic fallback and unified billing.

Change one string to switch providers #

The request shape never changes. Point your existing search and embeddings code at one base URL, set a model slug like exa/neural

or cohere/rerank-v3.5

, and let the router pick the endpoint. Add a provider

block to route by price, latency, or quality.

How routing works

curl https://searchrouter.ai/api/v1/search \
  -H "Authorization: Bearer $SR_API_KEY" \
  -d '{
    "model": "searchrouter/auto-search",
    "query": "best vector databases for RAG",
    "provider": { "sort": "quality" }
  }'

Use one search engine, or many at once #

You are not locked to a single provider per query. Fan out across several engines in one request, then dedupe and fuse the rankings so results multiple engines agree on rise to the top. Every result shows which engines returned it.

  • Mix providers. Combine Exa, Tavily, Brave and more in a single call, or let auto-search choose.
  • Choose search depth. Tune result count, pull full page content, and fuse across engines for recall.
  • One fused ranking. Reciprocal rank fusion with per-result source attribution and one bill.

Read about fan-out routing

Search is fragmenting into a multi-model world #

No single engine is best at every query. SearchRouter is the control plane for the many-model search era.

One schema, every provider

Web, neural, and keyword search, plus embeddings, reranking, extraction, and RAG answers - normalized behind a single request shape.

Routing & automatic fallback

Route by price, latency, or quality. When a provider errors or rate-limits, the next one takes the request - transparently.

Unified, pass-through billing

Prepaid credits across all providers at cost. Bring your own keys per provider, or use ours. One invoice, no per-vendor contracts.

Real usage rankings

See which search and retrieval models developers actually run, from live aggregate usage across the network.

Every search primitive, one canonical schema #

Start routing in minutes #

Create a key and run your first query against any provider. Demo mode works with zero upstream credentials.

── more in #ai-tools 4 stories · sorted by recency
── more on @searchrouter 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/show-hn-any-search-o…] indexed:0 read:2min 2026-06-17 ·