Skip to main content
The quickest way to get going with LanceDB is with the open source version, which is file-based and runs in-process (like SQLite). Let’s get started using Python in just a few steps!

​
1. Install LanceDB

Install LanceDB in your client SDK. 
pip install lancedb

​
2. Connect to a LanceDB database

Using LanceDB’s open source version is as simple as running the following import statement — no servers needed!

​
Connect to a database

Once you import LanceDB as a library, you can connect to a LanceDB database by specifying a local file path.

​
LanceDB Cloud or Enterprise versions

If you want a fully-managed solution, you can opt for LanceDB Cloud, which provides managed infrastructure, security, and automatic backups. Simply replace the local path with a remote uri that points to where your data is stored. 
db = lancedb.connect(
    uri="db://your-project-slug",
    api_key="your-api-key",
    region="us-east-1"
)
If you’re operating at enormous scale and are looking for more advanced use cases beyond just search, like feature engineering, model training and more, check out LanceDB Enterprise.

​
3. Obtain some data

LanceDB uses the notion of tables, where each row represents a record, and each column represents a field and/or its metadata. The simplest way to begin is to define a list of objects, where each object contains a vector field (list of floats) and optional fields for metadata. Let’s look at an example. We have the following records of characters in an adventure board game. The vector field is a list of floats. In the real world, these would contain hundreds of floating-point values and be generated via an embedding model, but the example below shows a simple version with just 3 values. 
data = [
    {"id": "1", "text": "knight", "vector": [0.9, 0.4, 0.8]},
    {"id": "2", "text": "ranger", "vector": [0.8, 0.4, 0.7]},
    {"id": "9", "text": "priest", "vector": [0.6, 0.2, 0.6]},
    {"id": "4", "text": "rogue", "vector": [0.7, 0.4, 0.7]},
]

​
5. Create a table

Next, let’s create a Table in LanceDB and ingest the data into it. If not provided explicitly, the table infers the schema from the data you provide. If the table already exists, you’ll get an error message. 
table = db.create_table("adventurers", data=data)
# If the table already exists, you'll get a ValueError: Table 'my_table' already exists
To overwrite the existing table, you can use the mode=overwrite parameter. 
table = db.create_table("adventurers", data=data, mode="overwrite")
# No ValueError!
LanceDB requires that you provide either data or a schema (e.g., PyArrow) during table creation. You can learn more about that in the “working with tables” page. Now, let’s perform a vector similarity search. The query vector should have the same dimensionality as your data vectors. The search returns the most similar vectors based on Euclidean distance. Our query is a vector that represents a warrior, which isn’t in the data we ingested. We’ll find the result that’s most similar to it! 
# Let's search for vectors similar to "warrior"
query_vector = [0.8, 0.3, 0.8]

# Ensure you run `pip install polars` beforehand
results = table.search(query_vector).limit(2).to_polars()
It’s often convenient to display the query results as a table. You can output the results of a vector search query as a Pandas or Polars DataFrame. The example below shows the output as a Polars DataFrame. It looks like the knight is the most similar adventurer to the warrior from our query!
idtextvector_distance
1knight[0.9, 0.4, 0.8]0.02
2ranger[0.8, 0.4, 0.7]0.02
If you prefer Pandas, you can use the to_pandas() method to display the results as a Pandas DataFrame. 
results = table.search(query_vector).limit(2).to_pandas()

​
7. Add data and run more queries

If you obtain more data, it’s simple to add it to an existing table. In the same script or a new one, you can connect to the LanceDB database, open an existing table, and use the table.add command.
Python
import lancedb

# Connect to an existing database
uri = "./ex_lancedb"
db = lancedb.connect(uri)

# Open the existing table that we created earlier
table = db.open_table("my_table")

more_data = [
    {"id": "7", "text": "mage", "vector": [0.6, 0.3, 0.4]},
    {"id": "8", "text": "bard", "vector": [0.3, 0.8, 0.4]},
]

# Add data to table
table.add(more_data)
To verify that our new data was added, we can run a different query that looks for adventurers similar to wizard.
Python
# Let's search for vectors similar to "wizard"
query_vector = [0.7, 0.3, 0.5]

results = table.search(query_vector).limit(2).to_polars()
print(results)
idtextvector_distance
7mage[0.6, 0.3, 0.4]0.02
9priest[0.6, 0.2, 0.6]0.03
The mage is the most magical of all our characters!

​
Temp

here’s a snippet: