Skip to main content
Search peer-reviewed papers, preprints, and scholarly datasets through one API - with full text, citations, and metadata. Built for literature reviews, research validation, and tracking developments across any field.

​
What you can do

Literature reviews

Synthesise scholarly sources across journals and preprint servers.

Citation discovery

Surface related work, references, DOIs, and citation counts.

Methods benchmarking

Compare methodologies, datasets, and evaluation metrics.

Trend analysis

Track topics, venues, and publication timelines over time.

​
Sources

SourceCoverageAccess
arXiv (valyu/valyu-arxiv)Physics, math, CS, quant-bio preprintsAll plans
PubMed (valyu/valyu-pubmed)Biomedical and life sciences literatureAll plans
bioRxiv (valyu/valyu-biorxiv)Life sciences preprintsSubscription
medRxiv (valyu/valyu-medrxiv)Health and clinical preprintsSubscription
ChemRxiv (valyu/valyu-chemrxiv)Chemistry and materials preprintsSubscription
CERN Open Data (cern-opendata)Particle physics datasetsSubscription
Wiley finance papers and booksLicensed finance and economics scholarshipSubscription
arXiv and PubMed are available on every plan, including free signup credits. bioRxiv, medRxiv, ChemRxiv, CERN, and licensed journals are unlocked by a subscription - which also lowers your cost per credit. See pricing and the full data catalog.

​
Quick start

from valyu import Valyu

valyu = Valyu("your-api-key-here")

# Let Valyu route to the right academic sources
response = valyu.search(
    "machine learning applications in quantitative finance",
    response_length="large"
)
print(response)

​
Target sources and dates

Use the academic preset to scope to scholarly sources, or pass dataset ids directly. Add start_date / end_date to focus on a window. 
response = valyu.search(
    "CRISPR gene editing therapeutic applications",
    included_sources=[
        "valyu/valyu-pubmed",
        "valyu/valyu-arxiv",
        "valyu/valyu-biorxiv",
        "valyu/valyu-medrxiv",
        "valyu/valyu-chemrxiv"
    ],
    start_date="2024-01-01",
    response_length="large",
    max_num_results=15
)
Use response_length="large" for academic work so you capture full methodology and results, not just abstracts.

​
Response format

Academic results include enhanced metadata - authors, citation, DOI, citation count, and references: 
{
  "results": [
    {
      "title": "Transformer Architecture for Protein Folding Prediction",
      "authors": ["Jane Smith", "John Doe"],
      "citation": "Smith, J., Doe, J. (2024). Nature Biotechnology, 42(3), 123-135",
      "citation_count": 45,
      "doi": "10.1038/s41587-024-12345",
      "publication_date": "2024-03-15",
      "content": "Detailed research content...",
      "references": "1. Previous work citation...",
      "source": "valyu/valyu-arxiv"
    }
  ]
}

​
Build a research agent

Wire Valyu into an AI research agent for academic literature reviews and validation.

Open in Cursor

​
Best practices

Use scholarly language and include methodology or field context: “peer-reviewed studies on neural plasticity”, “computational biology applications”, “meta-analysis of…”. This targets academic content over general web results.
Mix preprints (arXiv, bioRxiv, medRxiv, ChemRxiv) with peer-reviewed literature (PubMed) for interdisciplinary topics. Let Valyu auto-route, or pin specific datasets when you know what you need.
For ongoing literature monitoring, build a DeepResearch Workflow with a prompt template and scheduled runs, instead of re-issuing queries by hand.

​
Limitations

  • Some content sits behind paywalls or licensing restrictions.
  • Very recent work may not be indexed yet.
  • Source availability varies by discipline.
  • Results may favour English-language publications.