Hyperresearch
Deep research reports with every citation checked
What is Hyperresearch?
Hyperresearch is an AI deep research service that writes cited reports and checks each cited claim against its source page before delivery. It searches the open web and 90+ structured sources, including SEC filings, official statistics and PubMed, and every report comes with the full list of sources read and a table of the claims it relies on, each with its supporting quote. Every source it reads is saved to a vault in your workspace, and later runs search that vault before going back to the web.
Reports come in two sizes. A Light run reads about 20 to 50 sources and returns about 2,500 to 4,000 words in roughly 30 to 40 minutes, while a Deep run covers about 150 to 200 sources over 3 to 5 hours and adds a verification receipt recording which citations were checked and what the checker found. Hyperresearch notes that the Deep figures are pipeline targets and that neither size is guaranteed. For agents, Hyperresearch Answers is an MCP server with an ask tool that returns a short answer with checked quotes, or a not-found result instead of a guess, and a search tool that returns ranked sources. It connects to Claude, Cursor, VS Code, ChatGPT (on Business, Enterprise and Edu plans) and Windsurf, and the same calls are available through a REST API with webhooks.
There is no subscription. A Light report costs $9 and a Deep report costs $49, partial reruns cost 25 to 100 percent of the run price, and the first Deep run is free once you add a card. Answers calls start at 5 cents for a fast ask or a standard search, and every account gets $5 of free credit with no card needed. Enterprise options include a HIPAA business associate agreement, a private document vault and single sign-on. The underlying pipeline is also open source under the MIT license as a CLI that runs inside Claude Code or OpenAI Codex, built by founder Jordan Gibbs.
Score breakdown (out of 5)
What you would use it for
- →Market and competitor research with traceable citations
- →Giving coding and chat agents checked answers through MCP
- →Filings and literature research that builds a reusable source vault
Pros & Cons
👍 Pros
- ✓Pay per run with no subscription, seat licence or minimum
- ✓Returns not found instead of guessing when evidence is missing
- ✓Open-source CLI lets you run the same pipeline on your own agent plan
👎 Cons
- ✗Deep reports take 3 to 5 hours
- ✗The free Deep run requires adding a card
- ✗Per-report pricing adds up for frequent research
Key Features
- ✓ Cited research reports with each claim checked against its source
- ✓ Light ($9) and Deep ($49) report runs
- ✓ Claims table with supporting quotes and a full source list on every report
- ✓ Workspace vault that later runs search before the web
- ✓ Access to 90+ structured sources, including SEC filings and PubMed
- ✓ Answers MCP server with ask and search tools for agents
- ✓ REST API with scoped keys and webhooks
- ✓ Open-source MIT CLI for Claude Code and OpenAI Codex
Available on
Integrates with
Hyperresearch Pricing
✅ Hyperresearch has a free plan - no credit card required to start.
Open-source CLI
- ✓MIT license
- ✓Runs inside Claude Code or OpenAI Codex
- ✓Local searchable knowledge base with an MCP server
- ✓You pay for your own agent plan
Light report
- ✓About 20 to 50 sources
- ✓About 2,500 to 4,000 words
- ✓About 30 to 40 minutes
- ✓Cited claims checked against their sources before delivery
Deep report
- ✓About 150 to 200 sources
- ✓About 3 to 5 hours
- ✓Verification receipt for checked citations
- ✓First Deep run free after adding a card
Answers MCP
- ✓ask: 5 cents fast, 10 to 20 cents standard, 50 cents max
- ✓search: 5 cents standard, 15 cents max
- ✓$5 free credit, no card needed
- ✓Top-ups from $10
Enterprise
- ✓HIPAA business associate agreement
- ✓Private document vault
- ✓Retention or residency terms
- ✓Single sign-on and custom signed agreements
Developer resources
Related Tools
Claude is the best AI assistant for long-document work and research where accuracy matters. Its context window of up to 1M tokens and resistance to sycophancy make it the honest choice when you need a thinking partner rather than an agreeable assistant. Start on the free tier; the Pro plan at $20/month is worth it if you regularly hit the free usage limits, and Max ($100/month) or Ultra ($200/month) suit heavier daily use.
ChatGPT is the most capable all-in-one AI assistant available. Its combination of image generation, voice mode, code execution, web browsing, and a large custom GPT ecosystem makes it the default choice for users who want one tool that handles everything. The tradeoff is that it is rarely the deepest tool on any single task.
Jasper is the leading AI writing platform for marketing teams. Brand Voice, Campaigns, and workflow templates make it the most structured choice for organizations needing consistent output across multiple writers and channels. The pricing reflects an enterprise positioning; individuals who primarily need AI writing will find ChatGPT or Claude cheaper.
Perplexity is the best AI tool for research and fact-finding. Every answer cites its sources, making it easy to verify claims and dig deeper. For questions where accuracy and recency matter - news, technical details, market data - it outperforms ChatGPT on reliability. The free tier covers most use cases.
This page contains affiliate links. We may earn a commission at no extra cost to you. Learn more.