NewYour Exa or Tako SDK, up to 5× cheaper.

Use casesDaily briefings

Daily briefings, from sources your model can cite.

Newsletters, morning briefings and digests: the day’s coverage per section, the same story counted once, and verbatim excerpts so every sentence your model writes traces back to a source.

POST /v1/searchmode: "deep" · $5.60 per 1,000 queries

The problem

A summary is only as good as what it read

An AI-written briefing fails in familiar ways: it repeats the same wire story five times, misses the local outlet that broke it, or states something no source said.

We don’t write the briefing — your model does. We give it the day’s evidence: grouped, ranked, with verbatim quotes and links.

How it works

  • 01

    One request per section

    Up to five queries per section — “central bank rate decision”, “inflation data”, “fuel prices” — with days: 1.

  • 02

    The same story once

    dedupe: true folds the wire story that ten outlets ran into one result, with the duplicates attached and who published first.

  • 03

    Quotes, not paraphrase

    Deep mode returns highlights per article and essential: verbatim excerpts across outlets, the key facts your model builds on.

  • 04

    Your model writes

    Pass titles, sources, links and quotes to your model and ask it to cite them. Similar expands any story into full coverage.

API

The request

One request does it. The same parameters work over HTTP, with the SDKs and through the MCP server.

What it costs

Recommended · deep$5.60 per 1,000 queries

Deep reads 8 articles per query and returns highlights and essentials: the quotes a briefing is built on.

Also good · fast$1.40 per 1,000 queries

Fast ($1.40) for sections where headlines are enough, like a list of links.

An example month

A 10-section briefing, every day for 30 days

300 queries in deep mode ≈ $1.68

List prices from our pricing page. Each query in a multi-query request counts as one search; identical requests within 10 minutes are free.

curl https://api.typesearch.ai/v1/search \
-H "Authorization: Bearer $TYPESEARCH_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"query": ["central bank rate decision", "inflation data", "fuel prices"],
"mode": "deep",
"days": 1,
"max_results": 15,
"dedupe": true,
"essential": true
}'

What you get back

  • highlights and essential: verbatim excerpts, never generated text
  • duplicates: the same story from other outlets, and who published first
  • Title, URL, source and publication time for every citation
  • Results grouped by the query they answer

When to use something else

  • You want the API to write the briefing: we return evidence, not prose. An answer engine does that.
  • You need full article text in the newsletter: we return short excerpts that link to the original.

One result, trimmed

{
"title": "Central bank holds rate, signals cut in November",
"url": "https://diarioejemplo.example/economia/tasa",
"source": "Diario Ejemplo",
"published_at": "2026-09-25T18:30:00-03:00",
"score": 0.98,
"highlights": ["The board kept the policy rate unchanged and said a cut could come in November"],
"duplicates": [
{ "title": "Rate unchanged; November cut in view", "url": "https://wire.example/markets/rate", "source": "Example Wire" }
]
}

Questions

Does typesearch write summaries?

No, on purpose. We return verbatim excerpts, sources and links; your model writes. That way every sentence in the briefing can be traced back to what an outlet actually published.

Can I publish the excerpts in my newsletter?

Excerpts are short (up to 25 words, never from the first paragraph) and always come with the link to the original article. Quote them with attribution and send readers to the source.

What does a daily briefing cost?

300 queries in deep mode ≈ $1.68 at list price.

Build it this afternoon.

Get a key with $5 in free credit, run the request above and see what comes back.