Datasources

Search datasources

Find datasets by describing what you need in natural language, e.g. Wiley medical literature, FDA drug labels, insider transactions, CPI inflation. Ranking is semantic (embedding similarity) blended with exact-term matching, so both descriptive phrases and terse acronyms work.

Use the returned id values as included_sources on the Search API. Results are ranked best-first with a relevance_score in the range 0-1; a query matching nothing in the catalog returns an empty list rather than a weak guess.

Only datasets your key can search are returned (the same visibility rules as GET /v1/datasources).

get/v1/datasources/search

Request

  • Base URL: https://api.valyu.ai
  • URL: https://api.valyu.ai/v1/datasources/search
  • Auth: API key in header X-API-Key

Query parameters

qstring required

Natural-language description of the data you want.

limitinteger

Maximum number of datasets to return.

Response

Ranked datasets matching the query.

successboolean
querystring

The query as interpreted.

search_mode'semantic' | 'lexical'

semantic when embedding-based ranking was used; lexical when the full-text fallback served the request.

total_countinteger

Example response

{
  "results": [
    {
      "id": "valyu/valyu-arxiv",
      "name": "arXiv",
      "description": "Preprints in physics, mathematics, computer science, and more.",
      "category": "research",
      "type": "paper",
      "topics": [
        "physics",
        "mathematics",
        "computer science"
      ],
      "languages": [
        "en",
        "es",
        "fr"
      ],
      "source": "pubmed",
      "modality": [
        "text"
      ],
      "example_queries": [
        "CRISPR gene editing mechanisms",
        "quantum error correction methods"
      ],
      "pricing": {
        "cpm": 0.5
      },
      "update_frequency": "daily",
      "size": 2500000
    }
  ]
}

Changes

No recorded changes to this operation across all 1 revision of this API.