Web access API for agents whose Research endpoint runs multi-search reports with numbered, MLA, APA or Chicago citations.
Developers who want a single credit-metered API for the whole retrieval chain, since search, extract, crawl, map and a research agent sit behind one key and the credit cost of each is published. The Research endpoint fits report generation where the citation style matters: it takes numbered, MLA, APA or Chicago formats, accepts up to five attached text, Markdown or JSON files and cites any file it uses. Teams with procurement requirements get SOC 2, ISO 27001, GDPR and zero data retention claims plus a linked Trust Center.
Budgeting a per-report cost is awkward: Research pricing is dynamic, at 4 to 110 credits for model=mini and 15 to 250 for model=pro, so one report lands anywhere from roughly $0.03 to $2.00 at the $0.008 pay-as-you-go rate. Credits hard-stop rather than overflow, with the FAQ stating requests stop when they run out until the monthly reset or an upgrade, and per-plan rate limits are published only as 'higher' for Project and 'custom' for Enterprise. Domain control is partial: exclude_domains is a hard blocklist of up to 20, but include_domains is explicitly a soft preference, so other domains can still appear in a finished report.
Pros
Cons
You choose the format: citation_format accepts numbered (the default), mla, apa or chicago. Attached files are cited too, with up to five .txt, .md or .json files allowed as sources. Domain control is asymmetric: exclude_domains blocks up to 20 domains outright, while include_domains is a stated soft preference, so sources outside your list can still appear in the report.
Pricing is dynamic within a per-model range: 4 to 110 credits for model=mini and 15 to 250 credits for model=pro. At the $0.008 pay-as-you-go rate that is roughly $0.03 to $0.88 for mini and $0.12 to $2.00 for pro. Monthly plans lower the rate to between $0.0075 and $0.005 per credit depending on volume.
Yes. An August 2026 post reports 49.5% on SealQA-0, 55.7% on SealQA-Hard and 97.6% on SimpleQA Verified, with the method stated: 10 results per query, gpt-5.4 Mini as the reader at medium reasoning effort, gpt-4.1-mini as grader using the official SealQA prompt, identical across providers. The post names the competitors it beat but publishes none of their scores.
The Researcher plan is $0 with 1,000 API credits a month and no credit card required. That covers 1,000 basic searches or 500 advanced searches, or between 4 and 250 research reports depending on model and dynamic cost. Credits reset on the first of each month, and requests stop once they are exhausted until the reset or an upgrade. Students can apply for free access.
Facts sourced from the vendor’s own site, last verified 2026-10-10.
The Weekly Source
One email a week. New listings, repriced tools, and the ones we would skip - including this one if its pricing changes.
