Vector Store

Vector database for AI/ML embeddings powered by pgvector. Perfect for RAG applications, semantic search, and recommendation systems.

Features

• PostgreSQL-based with pgvector extension
• Support for multiple distance functions (L2, inner product, cosine)
• HNSW and IVFFlat indexing
• Hybrid search (vector + full-text)
• Automatic index optimization
• Compatible with popular embedding models

Plans

Plan	Dimensions	Vectors	Price
Starter	1536	100K	$18/mo
Standard	3072	1M	$54/mo
Pro	4096	10M	$135/mo

Quick Start

# Create a vector store
szc vector create my-vectors --plan starter --dimensions 1536

# Get connection string
szc vector info my-vectors

Usage with Python

import psycopg2
from pgvector.psycopg2 import register_vector

conn = psycopg2.connect("your-connection-string")
register_vector(conn)

# Create a table
cur = conn.cursor()
cur.execute("""
    CREATE TABLE documents (
        id SERIAL PRIMARY KEY,
        content TEXT,
        embedding vector(1536)
    )
""")

# Insert vectors
cur.execute(
    "INSERT INTO documents (content, embedding) VALUES (%s, %s)",
    ("Hello world", embedding)
)

# Search
cur.execute("""
    SELECT content, embedding <-> %s AS distance
    FROM documents
    ORDER BY distance
    LIMIT 5
""", (query_embedding,))

Indexing

Create an HNSW index for faster queries:

CREATE INDEX ON documents
USING hnsw (embedding vector_cosine_ops)
WITH (m = 16, ef_construction = 64);

OpenAI Integration

from openai import OpenAI

client = OpenAI()

def get_embedding(text):
    response = client.embeddings.create(
        model="text-embedding-3-small",
        input=text
    )
    return response.data[0].embedding