Search with vectors

Include vector embeddings in search results.

By default, search excludes vector data from results. To include the actual vector embeddings, set with_vectors=True.

Include vectors when you:

• Need embeddings for additional processing.
• Want to perform secondary similarity calculations.
• Require raw embedding data in your app.

Including vectors significantly increases response size, so only request them when necessary.

Python

import asyncio
from actian_vectorai import AsyncVectorAIClient
import random

async def main():
    # Connect to VectorAI DB server
    async with AsyncVectorAIClient("localhost:6574") as client:
        # Generate query vector
        query_vector = [random.gauss(0, 1) for _ in range(128)]
        
        # Include both payload and vector data
        results = await client.points.search(
            "my_collection",  # Collection name
            vector=query_vector,  # Query vector
            limit=10,  # Number of results
            with_payload=True,  # Include payload metadata
            with_vectors=True  # Include vector embeddings
        )
        
        # Display results
        for result in results:
            print(f"ID: {result.id}, Score: {result.score}")
            print(f"Payload: {result.payload}")
            if result.vectors:
                print(f"Vector dimensions: {len(result.vectors)}")

asyncio.run(main())

Each result includes these fields:

• id: The unique identifier of the matching point.
• score: Similarity score based on distance metric.
• payload: Full metadata dictionary for the point.
• vector: The complete vector embedding array.