The service currently runs on a single small machine: about 50 search requests/s and ~20 concurrent ask sessions across all users. Please keep concurrency modest (≤ 5 per client) and back off on 429 using Retry-After. Details →

Web

Overview

Agentic question answering over the open web. Only ask.

status
live
document id
url e.g. https://example.com/page
citation
[https://…]

Common calls

Ask the webPOST /v1/web/ask {"query": "…", "effort": "low"}

Askregistered key+ /stream

POST/v1/web/ask
Google search, reads pages, answers with URL citations. effort low / medium / high = 50 / 100 / 200 credits. Append /stream for NDJSON; the first event is always billing. A failure before the first answer_delta is refunded.

Parameters

ParameterTypeDescription
query REQUIRED string 1–2000 characters. Be specific; ask for numbers if you want numbers.
search_type search · scholar · news · images= search Google vertical
gl string= us Google country code; Chinese queries default to cn
hl string= en Search language; Chinese queries default to zh-cn
effort low · medium · high= low Reading rounds, and the price: 50 / 100 / 200 credits. Your plan caps it.
verbose bool= false Include the tool-call trace (tool_call / tool_result / thinking events when streaming).
top_k integer= 10 Speculative prefetch size.
stream_answer bool= true Streaming only: false emits one answer event instead of deltas.
max_rounds integer 0–8, overrides the effort preset.
force_answer_after number Seconds; hard cutover into the answer phase.
max_answer_tokens integer= 4096 Hitting it sets answer_truncated.
language string Answer language; defaults to the query’s.

Example

{ "answer": "… [https://…] …", "sources": [ … ], "stats": { "rounds": 1, "elapsed_s": 6.2, "answer_truncated": false }, "coverage"?: { … }, "meta": { "credits": { "charged": 50, … } } }
Citations are real — every id in answer exists in this domain (format [https://…]). sources is the retrieval set, not the citation list. answer_truncated means incomplete.