Skip to main content

​
Vector Databases

Deploy specialized vector databases for storing and querying embeddings for AI applications. Support for Milvus, Qdrant, Chroma, and Weaviate.

​
What are Vector Databases?

Vector databases are specialized databases designed to store and query high-dimensional vectors (embeddings) efficiently. They’re essential for AI applications like semantic search, recommendation systems, and RAG (Retrieval-Augmented Generation).

​
Supported Databases

​
Milvus

Open-source vector database built specifically for scalable similarity search and AI applications.

Learn More

Deploy Milvus

​
Qdrant

High-performance vector database and vector similarity search engine written in Rust.

Learn More

Deploy Qdrant

​
Chroma

AI-native open-source embedding database optimized for LLM applications.

Learn More

Deploy Chroma

​
Weaviate

Open-source vector database with a GraphQL API and built-in ML models.

Learn More

Deploy Weaviate

​
Use Cases

  • Semantic Search: Search by meaning, not keywords
  • RAG Applications: Retrieval-Augmented Generation for LLMs
  • Recommendation Systems: Find similar items
  • Image Search: Visual similarity search
  • Anomaly Detection: Find outliers in high-dimensional data
  • Deduplication: Find duplicate content

​
Getting Started

  1. Choose a vector database (Milvus, Qdrant, Chroma, or Weaviate)
  2. Create instance from your dashboard
  3. Configure collection and schema
  4. Get connection string and credentials
  5. Connect your AI application

​
Common Workflows

​
Store Embeddings

import openai
from pymilvus import connections, Collection

# Generate embedding
response = openai.Embedding.create(
    input="Your text here",
    model="text-embedding-ada-002"
)
embedding = response['data'][0]['embedding']

# Store in vector database
collection.insert([{"id": 1, "vector": embedding, "text": "Your text here"}])

​
Query Similar Vectors

# Search for similar embeddings
results = collection.search(
    data=[query_embedding],
    anns_field="vector",
    param={"metric_type": "L2", "params": {"nprobe": 10}},
    limit=10
)

​
Features

All Antryk vector databases include:
  • Scalability: Horizontal scaling for large datasets
  • High Availability: Multi-node clusters
  • Metadata Filtering: Filter by metadata attributes
  • Hybrid Search: Combine vector and keyword search
  • Monitoring: Performance metrics and query analytics
  • Backups: Automated backup and restore

​
Performance Optimization

  • Indexing: Built-in indexing for fast queries
  • Sharding: Distribute data across nodes
  • GPU Acceleration: Optional GPU support
  • Batch Operations: Efficient bulk inserts

​
Best Practices

  • Dimension Size: Choose appropriate embedding dimensions
  • Index Type: Select index based on your use case
  • Batch Size: Optimize batch insert sizes
  • Query Parameters: Tune search parameters for accuracy vs speed

Create Vector Database

Get started with vector databases