Skip to main content
Kinetica is a database with integrated support for vector similarity search. This notebooks goes over how to load documents from Kinetica 
pip install -qU langchain-kinetica
You must set the database connection in the following environment variables. If you are using a virtual environment you can set them in the .env file of the project:
  • KINETICA_URL: Database connection URL (e.g. http://localhost:9191)
  • KINETICA_USER: Database user
  • KINETICA_PASSWD: Secure password.
from gpudb import GPUdb

kdbc = GPUdb.get_connection()

2026-02-02 20:54:50.972 INFO     [GPUdb] Connected to Kinetica! (host=http://localhost:19191 api=7.2.3.3 server=7.2.3.5)
The following QUERY is an example which will not run; this needs to be substituted with a valid QUERY that will return data and the SCHEMA.TABLE combination must exist in Kinetica. 
from langchain_kinetica import KineticaLoader

QUERY = """select
    oid,
    object_name as text,
    creation_time
from ki_catalog.ki_objects
where schema_name = 'information_schema'
limit 10"""

kinetica_loader = KineticaLoader(
    kdbc=kdbc,
    query=QUERY,
    metadata_columns=["creation_time"],
)

kinetica_documents = kinetica_loader.load()
display(kinetica_documents)

[Document(metadata={'creation_time': 1769036399161}, page_content='oid: -263809000193198488\ntext: KEY_COLUMN_USAGE\ncreation_time: 1769036399161'),
    Document(metadata={'creation_time': 1769036399260}, page_content='oid: -6302080570668733378\ntext: KI_PERIODIC_OBJECTS\ncreation_time: 1769036399260'),
    Document(metadata={'creation_time': 1769036399402}, page_content='oid: 8620184385195410035\ntext: OBJECT_PRIVILEGES\ncreation_time: 1769036399402'),
    Document(metadata={'creation_time': 1769036399219}, page_content='oid: -582966432601743881\ntext: KI_HA_CONSUMERS\ncreation_time: 1769036399219'),
    Document(metadata={'creation_time': 1769036398744}, page_content='oid: -745341673129057292\ntext: COLUMNS\ncreation_time: 1769036398744'),
    Document(metadata={'creation_time': 1769036399143}, page_content='oid: 7168154014071633303\ntext: INFORMATION_SCHEMA_CATALOG_NAME\ncreation_time: 1769036399143'),
    Document(metadata={'creation_time': 1769036399352}, page_content='oid: 2994153253665268119\ntext: KI_QUERY_SPAN_METRICS_BY_SQL_STEP\ncreation_time: 1769036399352'),
    Document(metadata={'creation_time': 1769036399098}, page_content='oid: 2689064758811356754\ntext: INDEXES\ncreation_time: 1769036399098'),
    Document(metadata={'creation_time': 1769036398543}, page_content='oid: -7679099635043663749\ntext: APPLICABLE_ROLES\ncreation_time: 1769036398543'),
    Document(metadata={'creation_time': 1769036399434}, page_content='oid: 909900969452802786\ntext: ROLE_TABLE_GRANTS\ncreation_time: 1769036399434')]
Edit this page on GitHub or file an issue.
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.