VectorSearchRequest

Request schema for vector search with auto-embedding support.

Either query (text) or query_vector (pre-computed embeddings) must be provided. If both are provided, query_vector takes precedence.

query object

Text query to embed automatically (uses configured embedding service)

anyOf

• string
• null

string

Possible values: non-empty and <= 5000 characters

query_vector object

Pre-computed query embedding vector

anyOf

• number[]
• null

Array [

number

]

limitLimit (integer)

Maximum number of results to return

Possible values: >= 1 and <= 100

Default value: 10

offsetOffset (integer)

Number of results to skip (for pagination)

Possible values: >= 0

Default value: 0

score_threshold object

Minimum similarity score threshold (0.0 to 1.0)

anyOf

• number
• null

number

Possible values: >= 0 and <= 1

dataset_ids object

Filter by dataset IDs

anyOf

• integer[]
• null

Array [

integer

]

entry_ids object

Filter by entry IDs

anyOf

• integer[]
• null

Array [

integer

]

VectorSearchRequest

{
  "query": "string",
  "query_vector": [
    0
  ],
  "limit": 10,
  "offset": 0,
  "score_threshold": 0,
  "dataset_ids": [
    0
  ],
  "entry_ids": [
    0
  ]
}