Explain a vector search

POST /v1/tenants/{tenant}/vector/{table}/topk_explain

Inspect why a vector search behaved differentlyThe explanation reports the configuration used for this search.

Supply the query and settingsChoose dimension, metric, filters and execution options.
Run the explained searchRetrieve ranked hits with their configuration.
Compare config with expectationsInvestigate unexpected rankings using the actual settings.

Before you start

Use your database bearer token.

Store dense vectors in the collection before searching it. The example uses demo_vectors; scores depend on your data and distance metric.

Path parameters

Field	Type	Required
`tenant`	`string`	Required
`table`	`string`	Required

Headers

Field	Type	Required	Description
`x-oc-db-user`	`string`	Optional	Acting database user (role, grant and row-policy checks). Send with x-oc-db-key.
`x-oc-db-key`	`string`	Optional	API key of the x-oc-db-user user.

Request body

application/json

Limits and accepted values

Same body as the vector top-k search.
query must be finite numbers of length dim.
metric: cosine|dot|l2|manhattan.
filter: equality map {field: value} (explicit null is 400).
filter_expr: typed filter expression;
not combinable with a non-empty filter.
require_complete=true turns an underfilled filtered answer into 422.
mode: fast | high_recall (default).
engine: hnsw | bruteforce (internal tuning knob).
index: hnsw (default) | ivf | ivf_pq.
nprobe: 1..256 cells for ivf/ivf_pq.
refine_k_factor: ivf_pq exact re-rank over-fetch factor.

Field	Type	Required
`dim`	`integer`	Required
`embedder`	`string`	Optional
`filter`	`object`	Optional
`filter_expr`	`object`	Optional
`index`	`string`	Optional
`k`	`integer`	Required
`metric`	`string`	Optional
`mode`	`string`	Optional
`nprobe`	`integer`	Optional
`query`	`number[]`	Required

Fields in query

Field	Type	Required
`query`	`number`	Optional

Field	Type	Required
`refine_k_factor`	`integer`	Optional
`require_complete`	`boolean`	Optional

Read the result

hits is the ranked result and config shows the search configuration used. Use it to check dimensions, metric and execution choices when investigating unexpected rankings.

Response statuses

HTTP status reference

Status	What to do
200	OK
400	Correct the request before retrying.
401	Check the credential and Authorization header.
402	Check feature access or available credits.
403	Check tenant access, roles and permissions.
408	Reduce query work or adjust the timeout.
409	Resolve the conflict before retrying.
413	Split or reduce the request body.
415	Send the required Content-Type.
422	Correct the values or unsupported combination.
429	Honor Retry-After when present; back off.
499	The client cancelled the request.
500	Check the outcome before retrying a write.
501	Check deployment support for this operation.
503	Honor Retry-After when present; check write outcomes.

Request example

curl --include --request POST "${OC_BASE_URL}/v1/tenants/${OC_TENANT}/vector/demo_vectors/topk_explain" \
  --header "Authorization: Bearer ${OC_TOKEN}" \
  --header "Content-Type: application/json" \
  --data-binary '{
  "query": [
    0.1,
    0.2,
    0.3
  ],
  "dim": 3,
  "metric": "cosine",
  "k": 5
}'

Response example

Examples use synthetic values to show the response format.

Response status

{
  "config": {
    "dim": 3,
    "engine": "exact_scan",
    "index": "hnsw",
    "k": 5,
    "metric": "cosine"
  },
  "hits": [
    {
      "id": "v1",
      "score": 1
    },
    {
      "id": "v2",
      "score": 0.9331
    }
  ]
}

Illustrates a small-table exact-scan path; engine and tuning fields depend on the collection.

Download the OpenAPI specification for the complete response schemas.

Before you startCopy Anchor Link

Path parametersCopy Anchor Link

HeadersCopy Anchor Link

Request bodyCopy Anchor Link