Search vector store

POSThttps:/api.openai.com/v1/vector_stores/{vector_store_id}/search

Search a vector store for relevant chunks based on a query and file attributes filter.

Path parameters

  • vector_store_id
    string
    Required
    The ID of the vector store to search.

Request body

  • query
    string or array
    A query string for a search
    • query
      string
      Required
    • query
      array
      Required
      • items
        string
        A list of queries to search for.
  • rewrite_query
    boolean
    Defaults: false
    Whether to rewrite the natural language query for vector search.
  • max_num_results
    integer
    Defaults: 10
    The maximum number of results to return. This number should be between 1 and 50 inclusive.
  • filters
    object
    A filter to apply based on file attributes.
    • Comparison Filter
      object
      A filter used to compare a specified attribute key to a given value using a defined comparison operation.
      • type
        string
        Required
        Defaults: eq

        Specifies the comparison operator: eq, ne, gt, gte, lt, lte.

        • eq: equals
        • ne: not equal
        • gt: greater than
        • gte: greater than or equal
        • lt: less than
        • lte: less than or equal
        • eq
          string
        • ne
          string
        • gt
          string
        • gte
          string
        • lt
          string
        • lte
          string
      • key
        string
        Required
        The key to compare against the value.
      • value
        string or number or boolean
        The value to compare against the attribute key; supports string, number, or boolean types.
        • value
          string
          Required
        • value
          number
          Required
        • value
          boolean
          Required
    • Compound Filter
      object

      Combine multiple filters using and or or.

      • type
        string
        Required

        Type of operation: and or or.

        • and
          string
        • or
          string
      • filters
        array
        Required

        Array of filters to combine. Items can be ComparisonFilter or CompoundFilter.

        • Comparison Filter
          object
          A filter used to compare a specified attribute key to a given value using a defined comparison operation.
          • type
            string
            Required
            Defaults: eq

            Specifies the comparison operator: eq, ne, gt, gte, lt, lte.

            • eq: equals
            • ne: not equal
            • gt: greater than
            • gte: greater than or equal
            • lt: less than
            • lte: less than or equal
            • eq
              string
            • ne
              string
            • gt
              string
            • gte
              string
            • lt
              string
            • lte
              string
          • key
            string
            Required
            The key to compare against the value.
          • value
            string or number or boolean
            The value to compare against the attribute key; supports string, number, or boolean types.
            • value
              string
              Required
            • value
              number
              Required
            • value
              boolean
              Required
        • items
          object
  • ranking_options
    object
    Ranking options for search.
    • ranker
      string
      Defaults: auto
      • auto
        string
      • default-2024-11-15
        string
    • score_threshold
      number
      Defaults: 0

Response

A page of search results from the vector store.

Example request
1
curl -X POST \

2
https://api.openai.com/v1/vector_stores/vs_abc123/search \

3
-H "Authorization: Bearer $OPENAI_API_KEY" \

4
-H "Content-Type: application/json" \

5
-d '{"query": "What is the return policy?", "filters": {...}}'
Example response
1
{

2
  "object": "vector_store.search_results.page",

3
  "search_query": "What is the return policy?",

4
  "data": [

5
    {

6
      "file_id": "file_123",

7
      "filename": "document.pdf",

8
      "score": 0.95,

9
      "attributes": {

10
        "author": "John Doe",

11
        "date": "2023-01-01"

12
      },

13
      "content": [

14
        {

15
          "type": "text",

16
          "text": "Relevant chunk"

17
        }

18
      ]

19
    },

20
    {

21
      "file_id": "file_456",

22
      "filename": "notes.txt",

23
      "score": 0.89,

24
      "attributes": {

25
        "author": "Jane Smith",

26
        "date": "2023-01-02"

27
      },

28
      "content": [

29
        {

30
          "type": "text",

31
          "text": "Sample text content from the vector store."

32
        }

33
      ]

34
    }

35
  ],

36
  "has_more": false,

37
  "next_page": null

38
}
Built with