Skip to main content
POST
/
api
/
v1
/
vector
/
search
Search for nearest neighbors
curl --request POST \
  --url http://localhost:8080/api/v1/vector/search \
  --header 'Content-Type: application/protobuf+json' \
  --data '
{
  "vector": [
    123
  ],
  "k": 2,
  "nprobe": 123,
  "filter": {
    "and": "<array>",
    "or": "<array>"
  },
  "includeFields": [
    "<string>"
  ]
}
'
{
  "status": "success",
  "results": [
    {
      "score": 0.05,
      "vector": {
        "id": "doc-1",
        "attributes": {
          "vector": [
            1,
            2,
            3
          ],
          "category": "electronics"
        }
      }
    }
  ]
}

Body

Request body for searching nearest neighbors.

​
vector
number<float>[]
required

Query vector. Must have the same number of dimensions as the collection.

​
k
integer<uint32>
required

Number of nearest neighbors to return.

Required range: x >= 1
​
nprobe
integer<uint32>

Number of posting lists to search. Higher values improve recall at the cost of latency. Optional.

​
filter
object

A filter expression. Each filter object must contain exactly one operator key. Filters can be nested using and and or.

​
includeFields
string[]

List of attribute field names to include in results. When omitted, all attributes are returned.

Response

Search results ranked by similarity.

Response containing search results ranked by similarity.

​
status
string
required
Example:

"success"

​
results
object[]
required