Alibaba Cloud IQS Web Search compatible

The call, before and after

# Before
curl -X POST https://cloud-iqs.aliyuncs.com/search/unified \
  -H "Authorization: Bearer $IQS_API_KEY" -H "Content-Type: application/json" \
  -d '{"query": "Who founded Mistral AI and when?", "engineType": "Generic", "timeRange": "OneYear", "contents": {"mainText": true}}'

# After
curl -X POST https://api.specstream.ai/aliyun-iqs/search/unified \
  -H "Authorization: Bearer $SPECSTREAM_API_KEY" -H "Content-Type: application/json" \
  -d '{"query": "Who founded Mistral AI and when?", "engineType": "Generic", "timeRange": "OneYear", "contents": {"mainText": true}}'

The body and the parser stay. The host changes to api.specstream.ai/aliyun-iqs, and the key changes. The results come from the SpecStream Search API, in IQS's JSON.

Code on the previous version, GET /search/genericSearch with the key in X-API-Key, moves to https://api.specstream.ai/aliyun-iqs/search/genericSearch the same way.

Every parameter the route uses

Used in the search on UnifiedSearch: query (1 to 500 characters), engineType (Generic, GenericAdvanced, LiteAdvanced or Deep), timeRange (OneDay, OneWeek, OneMonth, OneYear or NoLimit), and in advancedParams: numResults (1 to 50), startPublishedDate and endPublishedDate as YYYY-MM-DD, which replace timeRange.

Used to shape the reply: in contents, mainText, markdownText and richMainBody set to true add the text of each page, summary adds the text of the page closest to the query, and rerankScore is on unless you send false.

On GenericSearch: query, timeRange, page (10 results a page), returnMainText and returnMarkdownText (on unless you send false), returnRichMainBody, enableRerank and advancedParams.

category, industry and locationInfo are accepted and not applied, so a client that sends them keeps working. An unknown engineType or timeRange is refused with HTTP 400 and the code InvalidParameter.

Every field that comes back

The reply has requestId, pageItems, sceneItems, queryContext with engineType, originalQuery.query, originalQuery.timeRange and rewrite.enabled, and searchInformation.searchTime in milliseconds. Each item of pageItems has title, link, snippet, publishedTime, mainText, markdownText, richMainBody, images, hostname, hostLogo, summary and rerankScore.

A text field you did not ask for is null. mainText holds up to 3,000 characters and richMainBody up to 12,000. images and sceneItems are empty lists, hostLogo is an empty string, and costCredits is not returned.

GenericSearch items add htmlTitle, displayLink, htmlSnippet, publishTime and score, and its searchInformation adds total.

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

A refusal keeps its HTTP status and carries IQS's requestId, code and message, with the same code and message again as errorCode and errorMessage for the IQS SDK: InvalidAccessKeyId.NotFound with 401, Retrieval.Arrears with 402 and Retrieval.Throttling.User with 429.

$1 per 1,000 searches, page text and summary included

A call of up to 10 results costs $0.001, with or without page text, on every engineType. Each started block of 10 more results adds $0.0012: numResults 20 costs $0.0022 and numResults 50 costs $0.0058. GenericAdvanced returns 50 results when you send no numResults, so send numResults 10 to pay $0.001.

The price does not move with your daily volume or with a QPS specification. 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.

Every call on this route 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 has no limit per key and no daily limit. A call that asks for page text reads the pages on the fast search, at the same price.

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.

Baidu AI Search on Qianfan

Baidu's search API on its own index, at ¥0.036 a call on the standard edition and ¥0.072 on the flagship edition, with 3 QPS by default and 1,500 free calls a month. SpecStream answers Baidu's POST /v2/ai_search/web_search as well, at https://api.specstream.ai/qianfan, for $0.001 a search of up to 10 results.

Tavily

A search API sold in credits at $0.008 each pay as you go: a basic search takes 1 credit and an advanced search 2. SpecStream answers Tavily's POST /search at https://api.specstream.ai/tavily for $0.001 a basic search and $0.005 an advanced one.

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. SpecStream answers Brave's call at https://api.specstream.ai/res/v1/web/search at $1 per 1,000, a fifth of Brave's rate, and has no limit per key.