A complete example for Multivector search is in this notebook

​
Multivector Type

LanceDB natively supports multivector data types, enabling advanced search scenarios where a single data item is represented by multiple embeddings (e.g., using models like ColBERT or CoLPali). In this framework, documents and queries are encoded as collections of contextualized vectors—precomputed for documents and indexed for queries.

Key features:

  • Indexing on multivector column: store and index multiple vectors per row.
  • Supporint query being a single vector or multiple vectors
  • Optimized search performance with XTR with improved recall.

Currently, only the cosine metric is supported for multivector search. The vector value type can be float16, float32, or float64.

​
Using Multivector in Python

Currently, multivector search is only support in our Python SDK. Below is an example of using multivector search in LanceDB:

Python
import lancedb
import numpy as np
import pyarrow as pa

# Connect to LanceDB Cloud
db = lancedb.connect(
  uri="db://your-project-slug",
  api_key="your-api-key",
  region="your-cloud-region"
)

# Define schema with multivector field
schema = pa.schema(
    [
        pa.field("id", pa.int64()),
        # float16, float32, and float64 are supported
        pa.field("vector", pa.list_(pa.list_(pa.float32(), 256))),
    ]
)

# Create data with multiple vectors per document
data = [
    {
        "id": i,
        "vector": np.random.random(size=(2, 256)).tolist(),  # Each document has 2 vectors
    }
    for i in range(1024)
]

# Create table
tbl = db.create_table("multivector_example", data=data, schema=schema)

# Create index - only cosine similarity is supported for multi-vectors
tbl.create_index(metric="cosine", vector_column_name="vector")

# Query with a single vector
query = np.random.random(256)
results_single = tbl.search(query).limit(5).to_pandas()

# Query with multiple vectors
query_multi = np.random.random(size=(2, 256))
results_multi = tbl.search(query_multi).limit(5).to_pandas()