# Web search for deep research agents: many queries, real sources > Deep research agents run dozens of searches per report. Plan queries, read the top pages as Markdown, dedupe and cite, from $1.20/1K searches. No KYC. - URL: https://llmwebsearch.com/use-cases/deep-research/ - Last updated: 2026-10-22 - Publisher: LLMWebSearch (https://llmwebsearch.com) A deep research agent does not run one search. It plans a set of queries, reads what comes back, notices the gaps and searches again. The quality of the report depends on how much real page text it sees along the way. **Quick answer — Which web search API suits a deep research agent?** A deep research agent needs a search API that returns page text, not just links, because it reads many sources per report. **LLMWebSearch** returns ranked results with the first 3 pages as Markdown for one credit per search. A report built from 40 searches uses 40 credits, so 100,000 searches covers 2,500 such reports. No KYC; pay in crypto. ## How does a deep research agent use search? The loop most designs share, whatever the model. 1. **Plan** — The model splits the question into four to eight short queries, each aimed at one part of the answer. 2. **Search and read** — Each query returns ranked results with the first pages already read, so the agent has text to work with at once. 3. **Keep the useful sources** — The agent drops pages it has already seen and pages that do not address the question. 4. **Look for gaps** — The model compares what it found with the question and writes new queries for what is still missing. 5. **Write with citations** — The final answer cites each claim with the number of its source, and says where sources disagree. ## How many searches does one report take? It depends on your design, and it is worth deciding on purpose. A light report might run two rounds of four queries: eight searches. A thorough one might run five rounds of eight: forty searches. More searches do not always mean a better report. The gain from each new round falls once the main sources are in hand. A stop rule helps: end the loop when a round adds fewer than two new sources, or when the budget is spent. With `read` at 3 or less, each search is one credit. A forty-search report therefore costs forty credits, whatever the length of the pages read. ## Why read pages instead of working from snippets? Because research is about details. Snippets give the gist of a page, but the figures, conditions and dates that make a report useful are in the body. A model that writes from snippets tends to fill gaps with guesses. Reading the pages also improves the next round of queries. The agent learns the terms the sources use, the names of laws or products, and the dates of events. Those words make sharper queries than the original question. LLMWebSearch reads the first pages of every search for you. Each result carries its position, title, URL, host, snippet and, when it was read, the page's main content as Markdown or plain text with its word count and publication date when known. ## How do I stop the agent from citing the same page twice? Key your source list by URL. Before adding a result, check whether its URL is already there. Different queries often return the same strong page, and you only need it once. Some pages appear under two URLs, for example with and without tracking parameters. Strip query strings that only track clicks before comparing. For news, the same story often runs on several outlets; keep the one with the earliest `published_at` or the most detail. Number the sources once, at the end, so the citations in the final answer match a clean list. ## When should I raise read above 3? When one query must cover a narrow subject in depth and the first three pages are not enough. Setting `read` between 4 and 10 makes that search 2 credits instead of 1. In most loops it is better value to run another query. A new query brings new pages from a different angle, while pages 4 to 10 of the same query often repeat the first three. Keep the higher setting for niche topics where few pages exist at all. ## How does this compare with ready-made deep research endpoints? Several vendors sell research as a single call, priced per task. As of October 22, 2026, Exa's deep search types cost $12 to $15 per 1,000 requests, and Exa retired its separate `/research` endpoint on April 1, 2026. You.com prices research calls from $12 (Lite) to $1,200 (Frontier) per 1,000. Parallel's Task API runs from $5 (Lite) to $2,400 (Ultra8x) per 1,000 tasks. Linkup's deep search is $50 per 1,000. These endpoints hide the loop, which is convenient. Building the loop yourself on a search API costs less per report at moderate depth, and you control the model, the prompts and the stop rule. The [step-by-step guide](https://llmwebsearch.com/guides/build-a-deep-research-agent/) has a working script of about 60 lines. ## Which pack suits a research workload? Count searches per report × reports per month. One credit pays for one search: up to 20 ranked results, with the pages of the first 3 read and returned as Markdown. Reading 4 to 10 pages makes it 2 credits. A call that ends in an error costs nothing. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Builder | 25,000 searches | $41.25 | $1.65/1K searches | | Production | 100,000 searches | $145 | $1.45/1K searches | | Scale | 500,000 searches | $650 | $1.30/1K searches | In every plan: Top 3 pages read as Markdown · Country, language, time and domain filters · Failed calls not charged · Credits never expire **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Money on your balance does not expire and buys search credits whenever you want. When a pack is cheaper than the minimum top-up, the part you paid only to reach that minimum comes back to you on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not returned unless the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Deep research: common questions ### Which model should write the report? Any model with a long enough context. The API returns sources, not a written answer: your own model reads the pages, writes the reply and cites the URLs, so you can use a large model for the final report and a small one for query planning. ### Can I limit research to trusted sites? Yes. Pass `include_domains` with up to 20 domains, or `exclude_domains` to drop sites you do not want cited. ### Can the agent research recent events only? Yes. Set `time_range` to `day`, `week`, `month` or `year`, and use `topic=news` for news coverage. ### How do I keep a research run within budget? Fix the number of rounds and queries per round in code, not in the prompt. Then the cost per report is known before it starts. ## Sources 1. [Exa pricing](https://exa.ai/pricing) — Exa (read 2026-10-22) 2. [You.com API pricing](https://you.com/pricing) — You.com (read 2026-10-22) 3. [Parallel pricing](https://parallel.ai/pricing) — Parallel Web Systems (read 2026-10-22) 4. [Linkup pricing](https://docs.linkup.so/pages/documentation/development/pricing) — Linkup (read 2026-10-22) --- LLMWebSearch: LLMWebSearch is a web search API for AI agents that returns ranked results with the content of the top pages in Markdown, paid in crypto with no ID upload. Prepaid credits from $1.20/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto.