Peer-reviewed science, one API call away.
Full-text papers, clinical trials, patents and structured databases for the AI your research teams rely on. Every answer comes back with its sources.
- Planning the search
- Searching PubMed, PMC full text, ClinicalTrials.gov
- Reading full text and results tables
- Writing a cited answer
Loss-of-function variants in PCSK9 lower LDL cholesterol and are associated with a substantially lower risk of coronary heart disease 1. Gain-of-function mutations cause familial hypercholesterolaemia, which fixes the direction of effect 2.
Evolocumab reduced major cardiovascular events on top of statins in the FOURIER outcomes trial 3.
on DRACO, the long-form research benchmark, across deep research, medicine and scholarly research.
Biology, chemistry and medicine literature synthesis
Hard medical questions, ahead of Google, Exa and Parallel
Web, academic, biomedical, legal and financial
Fields
Choose your field.
Same API, same sources. Each page is written for the people in that field.
Evidence for every stage, from target to label.
Clinical and regulatory evidence for design, submission and surveillance.
Reported properties pulled from tables, with their conditions attached.
arXiv, patents and CERN open data for hardware and physical R&D.
Full-text search in your terminal, for reviews, papers and lab agents.
Seven data layers and 100+ sources, filterable by field.
Why full text
Science needs sources, not generic search.

Full text, not abstracts.
Methods, results and supplementary tables, not the 250-word summary.
A source for every claim.
Answers return with citations to the papers, trials and patents they came from.
Licensed for commercial use.
Proprietary journals, cleared for use inside your products and internal tools.
Built for agents.
REST, MCP and SDKs, so your agent searches the record like any other tool.
Full text
Drag to see what an abstract leaves out.
Most AI tools read the abstract. The methods, results tables and supplements are where the evidence is.
Products
Two ways to use the scientific record.
Search when your agent needs the evidence. DeepResearch when you need the whole review. Both return sources.
Cited full text for your agent, in one call.
Ask in plain language and get passages, tables and figures back from papers, trials, labels and patents, each linked to its source. For agents that reason over the evidence themselves.
How Search works →A literature review, run for you.
Give it a research question. It plans the searches, reads the full text, checks the claims and returns a cited report. Heavy mode scores 72.7% on DRACO.
How DeepResearch works →Set up Valyu by following https://www.valyu.ai/agents.mdGuide →Also works with LangChain·LlamaIndex·Vercel AI SDK·Anthropic·OpenAI·Google·AWS Bedrock·n8n·CLI·Claude Code plugin
Benchmarks
State of the art across the science that matters.
ScholarQA-Multi
Accuracy on scientific-literature synthesis across biology, chemistry and medicine.
Use cases
The work it does.
Discovery & R&D
Find, rank and test ideas against everything already published.
Clinical & regulatory
Trials, clearances, labels and precedents, read alongside the literature.
Evidence synthesis
Screen, extract and summarise a literature, with the audit trail kept.
Diligence & scouting
Check the science, trial record and IP behind an asset or technology.
Writing & review
Drafts and checks where every claim carries its source.
Storke cut clinical evaluation timelines in half — grounding every assessment in the primary literature instead of hand-searching it.
FAQ
Questions researchers ask.
Is the content licensed for commercial use?
How fresh is the data?
Can I restrict answers to peer-reviewed sources?
Does it work with my agent framework?
Which plan includes the science sources?
Enterprise on Valyu.
Private indexes on your firm's internal data, enterprise controls, and engineers who ship with you.
Talk to our team
SOC 2, Zero Data Retention, SSO with SAML or OIDC, scoped tenant isolation. US, EU, and UK deployment options, or dedicated in your VPC.

State-of-the-art retrieval on internal data: lab notebooks, internal reports, trial data, research archives. Agents run on your own research and every Valyu source, in one call.

When the API needs to go further, our engineers ship with you. Custom DeepResearch agents and vertical indexes tuned to your field.
Everything your scientists read, behind one API call.
- 37M+PubMed papers
- 500K+clinical trials
- 150K+drug labels
- 100M+compounds
- 12M+patents

