Elicit alternatives in 2026

One table of everything on the question, grey literature included

Grey literature sits in no single index. On SpecStream you state the question as your protocol does, with the conditions every row must meet and the facts to bring back for each one. It can be studies, reports, trials or datasets: "working papers and agency reports since 2020 on cash transfers and school attendance in sub-Saharan Africa, with the publisher, the country and the sample size", or "registered clinical trials of psilocybin for depression, with the sponsor, the phase and the status".

SpecStream reads the request back as parts you can edit before anything runs, then searches the live web, reads each page and joins what it finds into one record per paper, report or trial. Every row meets the conditions you wrote.

Your methods section can say how it was searched

The request is kept as written and can rerun. Each value carries the URL of the page it was read on and the time it was read, so every row has a URL to cite. For any row, get_find_decisions returns each condition's answer with its reason, the value, the source and the time, and replay_find_decisions reads the saved record again at no charge: same record, same answer.

You pay for rows, with no seats

Elicit bills each user by the month: $588 a year on Pro and $2,748 a year on Scale. SpecStream bills the rows that meet your request: 5¢ for a row with up to 5 facts, 15¢ for up to 15 facts and one set of related records, and $1 for a deep row with no limit. One Pro seat's $588 is 11,760 rows at 5¢. The searches and pages behind each row are not billed, and a run that finds nothing costs nothing.

Every request gets a quote of the most it can cost, good for 30 minutes. Nothing runs until you approve it, the charge never goes past that amount, and whatever the run does not use goes back to your balance. The full rates are on pricing.

A living review stays up to date

Save the request and it reruns hourly to yearly. Each row comes back labeled changed, unchanged or no longer matching. A changed or new row costs half its rate and an unchanged row costs 1¢, so rechecking 1,000 rows that did not change costs $10. New matches land in Google Sheets, Excel, Airtable, Notion, Slack, Teams or email, or in your agent through the MCP server. Rows you already own are never charged again.

Research funding data and published benchmark scores

SpecStream holds 126,078 NIH awards ($185.3B) and 50,725 NSF awards ($31B) to 8,035 organizations, read Nov 9, 2026, at specstream.ai/data/research-funding. Ask for your own cut: the awards in your field, by institution, year or amount.

SpecStream's Search API is the most accurate web search API in the world. It scores 97.9 F-score on all 1,000 official SimpleQA Verified questions (978 correct), and 64.0 item F1 and 43.4 row F1 on all 200 official WideSearch tasks. The Extract API scores 0.908 F1 on WCXB. Scores and method are on benchmarks.

SpecStream

The live web data engine for teams and agents. One request in plain words returns the papers, reports, trials and pages that answer a research question as structured rows at 5¢, 15¢ or $1 each, with the source URL and read time on every value. New accounts get $10 of credit, no card.

Scite

Scite sells its research platform by plan. Basic is $12 a month billed yearly, with 250 MCP credits a month and collections of up to 1,000 papers. Pro is $40 a month billed yearly, with 2,500 MCP credits a month and API access, and Enterprise is priced by quote. SpecStream charges each delivered row at one rate on every plan, under a maximum you approve before the run.

Litmaps

Litmaps draws maps of related articles. Pro is listed at $10 a month on annual billing, $120 a year, and that price requires an academic email address. Team is sold by quote. SpecStream returns the question's papers and reports as rows with the facts you asked for on each, and anyone can sign up at the same rates.

Semantic Scholar API

An API to the Semantic Scholar paper index, with no price on its page. Calls without a key share 1,000 requests per second among all unauthenticated users, and a new API key starts at 1 request per second on all endpoints. SpecStream's Search API answers from the live web at $5 per 1,000 searches, and Deep Search returns the whole table from one request.

ResearchRabbit

A paper discovery tool with access to over 320 million academic papers, free to sign up. SpecStream reads the live web, so the reports, registers and agency pages that no paper index holds arrive in the same table as the papers, each value with its source URL and read time.