Exa Contents compatible

The call, before and after

# Before
from exa_py import Exa
exa = Exa(api_key="...")

# After
exa = Exa(api_key=SPECSTREAM_KEY, base_url="https://api.specstream.ai/exa")

reply = exa.get_contents(["https://en.wikipedia.org/wiki/Web_scraping"], text=True)
for result in reply.results:
    print(result.title, result.url, result.text[:300])

exa-py 2.28.0 joins base_url to /contents, so get_contents and search both reach SpecStream. With no content option exa-py sends text.maxCharacters, and the route returns the text cut at that length.

Apps and SDKs that take the address as a setting

Each line was run on production on November 10, 2026 through that client's own request and reply code. Put your SpecStream key wherever the client takes its key.

Khoj: in the Django admin, set api_url on the Exa WebScraper row to https://api.specstream.ai/exa. EXA_API_URL only fills that row when it is first made.

What the route reads, and what it refuses

Read and applied: urls as one string or a list, ids in their place, text as true, and text.maxCharacters. A host that does not resolve is listed in statuses with status error and its tag.

Refused with Exa's error body: more than 20 pages and urls sent with ids get 400, and a wrong key gets 401. Every parameter was measured on production on Nov 10, 2026 with exa-py 2.28.0, and the full pass table lists each one and its result.

Every field that comes back

The reply has requestId, results[], statuses[] with id, status, source and error.tag, costDollars.total with the call's price, and searchTime in milliseconds. Each result has id, url, title, publishedDate and text.

The route does not return context, a result's author, highlights, highlightScores, summary, image, favicon, subpages, extras or entities, or a status's error.httpStatusCode.

$0.80 per 1,000 pages

Exa bills $1 per 1,000 pages for each content type a call asks for. On SpecStream every page that gives text costs $0.0008, and a page with no text is not charged. costDollars.total on each reply is what the call cost. A busy call gets 429 with Retry-After and charges nothing.

0.908 F1 on WCXB, and the search scores on the same key

The Extract API that answers this route scores 0.908 F1 on WCXB. The Search API on the same key scores 97.9 F-score on all 1,000 official SimpleQA Verified questions, 978 of them correct, and 64.0 item F1 and 43.4 row F1 on all 200 official WideSearch tasks. Every score and its method is on the benchmarks page.

You decide what is kept, on each key

Create the key with zero_retention and a page read made with it keeps no request text and no answer after the reply. spend_cap_cents is the most the key can spend.

curl -X POST https://api.specstream.ai/v1/account/keys \
  -H "Authorization: Bearer $SPECSTREAM_API_KEY" -H "Content-Type: application/json" \
  -d '{"name": "production", "scopes": ["account:read", "finds:spend"], "spend_cap_cents": 500000, "zero_retention": true}'

Search, pages and rows on one key

The same key searches the live web with the Search API at $5 per 1,000 searches of up to 10 results. Deep Search returns every match for a request in plain words as structured rows, at 5¢, 15¢ or $1 a row, with a source URL on every value. The most a run can cost is approved before it starts. The full comparison is on the Exa alternative page.

Tavily Extract

Every 5 successful URL extractions cost 1 credit on basic, at $0.008 a credit on pay as you go, in Tavily's own reply. SpecStream reads 1,000 pages for $0.80 and answers the Exa call your code already sends.

Firecrawl

A scrape API at 1 credit per page, with the Hobby plan at $19 a month and 1,000 free credits a month. SpecStream needs no plan and keeps Exa's reply shape.

Jina Reader

Page reads billed in tokens by content length, with 10M free tokens on a new key and no dollar rate on its Reader page. SpecStream charges $0.0008 a page, whatever its length.