Web

Overview

General agentic search over the open web, for what was never going to be in a paper: current events, products, pricing, documentation. Google plus cached page bodies — the service never fetches live. Same request shape, same NDJSON protocol and the same agentic quota as Papers.

Ask the webregistered key+ /stream

POST/web/agent/search

Same request shape as Ask arXiv, answered from Google results plus cached page bodies. Answers with markdown links to real URLs; each source carries read: true|false. Shares the agentic quota.

CLIdeepxiv ask --web "…"

Parameters

ParameterTypeDescription
query string REQUIRED 1–2000 characters. Be specific; ask for numbers if you want numbers.
effort default | high | xhigh = default Gather rounds ceiling: 1–2 / 3 / 4–5.
verbose bool = false Include the tool-call trace (and tool_call / tool_result / thinking events when streaming).
search_type search | scholar | news | images = search Google vertical.
gl string = us Google country code. Chinese queries auto-switch to cn.
hl string = en Search language. Chinese queries auto-switch to zh-cn.
stream_answer bool = true Streaming endpoint only. false emits one answer event instead of deltas.
max_rounds integer OPTIONAL 0–8. Overrides the effort preset. 0 answers straight off the prefetch — first token ≈1.3s.
force_answer_after number OPTIONAL 0.5–30s. Hard cutover into the answer phase.
max_answer_tokens integer = 4096 256–16384. Hitting it sets answer_truncated.
language string OPTIONAL Answer language. Defaults to the query’s language.
Web evidence has two strengths. Only cached page bodies are read — never a live fetch — so an uncached page contributes just its snippet. Surface read so your agent can qualify weaker claims. Chinese sites and news pages are cached less often.
Web is slower than arXiv: cache misses cost 1.7–4.3s upstream. First-token target is 10s, not 5s.

Raw web search

POST/websearch

Organic results from the upstream search API, no reading, no answer. Costs 50 general requests per call. Prefer Ask the web unless you want the result list itself.

Parameters

ParameterTypeDescription
query string REQUIRED Search query.
search_type scholar | web = scholar scholar or web.
page integer = 1 Result page.
use_cache bool = true Serve from cache when available.

50 general requests per call