██╗ ██╗███████╗ ██████╗██╗ ██╗██╗ ██╗
██║ ██║██╔════╝██╔════╝██║ ██╔╝╚██╗ ██╔╝
██║ ██║█████╗ ██║ █████╔╝ ╚████╔╝
╚██╗ ██╔╝██╔══╝ ██║ ██╔═██╗ ╚██╔╝
╚████╔╝ ███████╗╚██████╗██║ ██╗ ██║
╚═══╝ ╚══════╝ ╚═════╝╚═╝ ╚═╝ ╚═╝
Semantic search for agents
| Feature | Description |
|---|---|
| Search | Find by meaning, not keywords. State-of-the-art Voyage AI embeddings. |
| Corpus | 2M+ papers from arXiv, Semantic Scholar, DBLP, OpenAlex, Crossref, and more. |
| Speed | Turbopuffer vector database. Sub-100ms p99 latency. |
| Custom | Index your own documents. Private collections. (Pro+) |
| API | Simple REST API. JSON responses. Bearer auth. |
Indexed Sources
35 base sources continuously indexed. Filter by source in your API calls with the sources parameter. Derived per-topic or per-venue sources (e.g. openalex_pacmpl, dblp_pldi, semantic_scholar_pl, semantic_scholar_author_joshua_tenenbaum, crossref_compilers, pubmed_neuroscience, osf_psyarxiv) are also available for narrower filters.
| Source | Description |
|---|---|
arxiv |
Open-access preprints in physics, math, CS, and more. 2M+ papers. |
semantic_scholar |
AI-powered academic search. Citations, abstracts, full-text links. |
dblp |
Computer science bibliography. Conference and journal papers. |
openalex |
Open catalog of scholarly works, authors, institutions. |
crossref |
Crossref metadata across journals, proceedings, and DOIs. |
pubmed |
Biomedical and neuroscience papers via NCBI. |
biorxiv |
Biology and neuroscience preprints. |
medrxiv |
Medical and health preprints. |
osf_preprints |
Preprints across OSF communities. |
openreview |
Open peer review. ICLR, NeurIPS, and ML conference papers. |
acl_anthology |
NLP and computational linguistics papers from ACL events. |
iacr |
International Association for Cryptologic Research. ePrint archive. |
acm |
ACM Digital Library. Computing research and publications. |
zenodo |
CERN's open repository. Research data, software, papers. |
enrich_dblp |
DBLP enrichment pass for author and venue metadata. |
| Source | Description |
|---|---|
nlab |
Category theory, homotopy theory, and higher mathematics wiki. |
| Source | Description |
|---|---|
hacker_news |
Y Combinator's tech news. Top stories and discussions. |
hn_algolia |
HN Algolia search backfill for long-tail threads. |
lobsters |
Computing-focused link aggregator. Curated tech discussion. |
github |
Repository READMEs, documentation, issues, and code. |
lesswrong |
Rationality, AI alignment, and effective altruism essays. |
ea_forum |
Effective altruism discussions, research, and community posts. |
community_archive |
Community-curated long-form posts and discussions. |
Sources are refreshed via Oban cron jobs (cadence varies by source). Custom source requests? Let us know.
Pricing
Need more? Contact us for Enterprise plans.
Documentation
Search the corpus with a natural language query.
Parameters
| query | Search query (required) |
| limit | Results to return (default: 10, max: 100) |
| sources | Filter by source (optional). Accepts base sources listed above and derived overrides like openalex_pacmpl or dblp_pldi. Base sources include all derived overrides. |
Returns ranked results with scores and metadata.
Include your API key in the Authorization header:
API keys start with vky_sk_.
Create keys in your dashboard.
| Plan | Requests/min | Monthly Quota |
|---|---|---|
| free | 10 | 1,000 |
| pro | 60 | 50,000 |
| team | 300 | 500,000 |
Rate limit headers: X-RateLimit-Remaining,
X-RateLimit-Reset