Migrate from Tavily

The call, before and after

# Before
curl -X POST https://api.tavily.com/search \
  -H "Authorization: Bearer $TAVILY_API_KEY" -H "Content-Type: application/json" \
  -d '{"query": "Who founded Mistral AI and when?", "max_results": 5}'

# After
curl -X POST https://api.specstream.ai/tavily/search \
  -H "Authorization: Bearer $SPECSTREAM_API_KEY" -H "Content-Type: application/json" \
  -d '{"query": "Who founded Mistral AI and when?", "max_results": 5}'

The body and the parser stay. The base address gains /tavily, and the key changes. A call with no max_results returns 10 results, as Tavily's does.

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.

LangChain, langchain-tavily 0.2.18: pass api_base_url="https://api.specstream.ai/tavily" to TavilySearch.

Every parameter, and what the route does with it

Read and used in the search: query, max_results (0 to 20), include_domains and exclude_domains (a list, or names apart by commas), time_range (day, d, week, w, month, m, year, y), topic news with days, start_date and end_date.

Read to choose the search: search_depth advanced, include_answer, include_raw_content and a max_results above 10 run the default search; every other call runs the fast search. include_raw_content adds the page text. The price is the same on both.

Read to shape the reply: include_answer (true, basic or advanced), include_raw_content (true, markdown or text) and include_usage. Accepted and not used: country, include_images, max_age_hours and the rest, so a client that sends them is never refused.

Every field that comes back

The reply has query, follow_up_questions (null), answer, images (an empty list), results, response_time in seconds and request_id, plus usage {"credits": 1} when you send include_usage. answer is filled when you ask for it and the search has one, and null otherwise.

Each result has title, url, content, score and raw_content, and published_date when the page has a date. score follows the rank order from the top result down, so code that sorts by it keeps the order. raw_content is the text of the page closest to your query, up to about 3,500 characters, or null when you did not ask for it. The route does not return favicon or images.

One field is added: specstream, with what the Search API returns beyond Tavily's fields, including the call's price_usd. A parser that reads Tavily's fields never sees it.

$5 per 1,000 searches, basic or advanced

On Tavily pay as you go, 1,000 basic searches cost $8 and 1,000 advanced searches $16. On SpecStream every call of up to 10 results costs $0.005, with page text and the answer included, and max_results 20 costs $0.011. A call stopped by a 429 charges nothing, and credit you buy never expires.

The most accurate web search API, in about one second

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

A basic Tavily call runs the fast search. Measured on production on November 9, 2026, its median was 0.82 s at 1 search a second and 1.07 s at 20 a second, the 95th percentile stayed between 1.09 and 1.73 s, and it answered 718 of 720. It takes up to 8,800 searches a minute on one account.

You decide what is kept, on each key

Create the key with zero_retention and a search made with it keeps no request text and no answer after the reply: it is not in the history, its answer is not reused for another call, and its log lines carry neither. 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 reads up to 20 pages or documents a call with the Extract API at $0.80 per 1,000 pages, and a page with no text is not charged. 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 Tavily alternative page.

Exa

A search API on Exa's own index, at $7 per 1,000 requests on auto and $4 on lite for up to 10 results, with its own request and reply. On SpecStream a Tavily call of up to 10 results costs $0.005 and needs no new code.

Brave Search API

A search API on Brave's own index at $5 per 1,000 requests, with $5 in free credits every month and 50 queries per second, in Brave's own reply shape. SpecStream costs the same per call and answers in Tavily's shape.

Serper

Google results as JSON, sold in credit packs from $1.00 per 1,000 for 50,000 credits, and the credits are valid for 6 months. Credit you buy on SpecStream never expires, and the call your code sends to Tavily keeps working.

Parallel

A Search API that starts at $1 per 1,000 requests, with 600 Search calls a minute on the Developer plan. Its migration guide maps Tavily calls to Parallel's own request. SpecStream answers the Tavily call as it is, and takes up to 8,800 searches a minute on one account.