Pinecone Query
CertifiedQuery a Pinecone index by vector similarity
Pinecone Query
Query a Pinecone index by vector similarity
Queries a Pinecone index for the top-K most similar vectors.
Provide either a vector (dense values) or a vectorId (query by existing vector ID).
Exactly one of these two must be set.
Results can be returned inline (FETCH / FETCH_ONE) or stored as an ION file (STORE).
type: io.kestra.plugin.pinecone.QueryExamples
Query top-5 vectors by a dense vector.
id: pinecone_query
namespace: company.team
tasks:
- id: query
type: io.kestra.plugin.pinecone.Query
apiKey: "{{ secret('PINECONE_API_KEY') }}"
indexName: my-embeddings
namespace: my-namespace
topK: 5
vector: [0.1, 0.2, 0.3]
includeMetadata: true
fetchType: FETCH
Query and store results to Kestra internal storage.
id: pinecone_query_store
namespace: company.team
tasks:
- id: query
type: io.kestra.plugin.pinecone.Query
apiKey: "{{ secret('PINECONE_API_KEY') }}"
indexName: my-embeddings
topK: 100
vector: [0.1, 0.2, 0.3]
fetchType: STORE
Properties
apiKey *string
Pinecone API key
indexName *string
Index name
Name of the Pinecone index to operate on.
topK *integerstring
Number of results to return
assets
Assets this task consumes as inputs or produces as outputs, for lineage tracking and the asset graph (Enterprise Edition). A flow declaring this property on a task is rejected in the open-source edition.
io.kestra.core.models.assets.AssetsDeclaration
IGNOREFAILWARNAsset failure behavior
Behavior applied to the task state when a declared asset fails to render, emit, or be persisted (e.g. a lock conflict): FAIL escalates it to FAILED, WARN (default) warns it if it would otherwise succeed, IGNORE leaves the state untouched.
Whether to auto-register assets referenced dynamically at runtime that are not statically declared in inputs or outputs.
The assets consumed as inputs.
io.kestra.core.models.assets.AssetIdentifier
1The assets produced as outputs.
io.kestra.plugin.ee.assets.Dataset
1150{}1150io.kestra.plugin.ee.assets.File
1150{}1150io.kestra.plugin.ee.assets.Table
1150{}1150io.kestra.plugin.ee.assets.VM
1150{}1150io.kestra.core.models.assets.External
1150{}1150io.kestra.core.models.assets.Custom
11501Custom asset type
{}1150fetchType string
FETCHFETCHFETCH_ONESTOREFetch type
Controls how results are returned. FETCH returns all rows inline, FETCH_ONE returns only the first row, STORE writes all rows to Kestra storage as an ION file.
filter object
Metadata filter
Optional filter map applied to restrict matches by metadata fields.
host string
Control-plane host override
Optional URL override for the Pinecone control-plane API. Defaults to https://api.pinecone.io when unset. Set this to point to a local emulator (e.g. http://localhost: 5080).
includeMetadata booleanstring
falseInclude metadata in the response
includeValues booleanstring
falseInclude vector values in the response
namespace string
Namespace to query
tlsEnabled booleanstring
trueEnable TLS for gRPC data-plane connections
Set to false when connecting to a local emulator that does not support TLS. Defaults to true (production Pinecone always requires TLS).
vector array
Query vector (dense values)
Mutually exclusive with vectorId.
vectorId string
ID of an existing vector to use as the query
Mutually exclusive with vector.
Outputs
row object
Single result row (FETCH_ONE mode)
rows array
All result rows (FETCH mode)
size integer
Number of matches returned
uri string
uriURI of the ION file containing all rows (STORE mode)
