Create Your First Index

Create a ParadeDB index over the mock_items table from Configure Your Environment. The same index supports text search, vector search, filtering, sorting, and aggregations in the next page.

CREATE INDEX search_idx ON mock_items
USING paradedb (
  id,
  description,
  embedding vector_cosine_ops,
  category,
  rating,
  in_stock,
  created_at,
  last_updated_date,
  latest_available_time,
  metadata,
  weight_range
);

import { sql } from "drizzle-orm";

await db.execute(sql`
  CREATE INDEX search_idx ON mock_items
  USING paradedb (
    id,
    description,
    embedding vector_cosine_ops,
    category,
    rating,
    in_stock,
    created_at,
    last_updated_date,
    latest_available_time,
    metadata,
    weight_range
  )
`);

from django.db import connection

with connection.cursor() as cursor:
    cursor.execute("""
        CREATE INDEX search_idx ON mock_items
        USING paradedb (
          id,
          description,
          embedding vector_cosine_ops,
          category,
          rating,
          in_stock,
          created_at,
          last_updated_date,
          latest_available_time,
          metadata,
          weight_range
        )
    """)

from sqlalchemy import text

with engine.begin() as conn:
    conn.execute(text("""
        CREATE INDEX search_idx ON mock_items
        USING paradedb (
          id,
          description,
          embedding vector_cosine_ops,
          category,
          rating,
          in_stock,
          created_at,
          last_updated_date,
          latest_available_time,
          metadata,
          weight_range
        )
    """))

ActiveRecord::Base.connection.execute <<~SQL
  CREATE INDEX search_idx ON mock_items
  USING paradedb (
    id,
    description,
    embedding vector_cosine_ops,
    category,
    rating,
    in_stock,
    created_at,
    last_updated_date,
    latest_available_time,
    metadata,
    weight_range
  );
SQL

await dbContext.Database.ExecuteSqlRawAsync("""
    CREATE INDEX search_idx ON mock_items
    USING paradedb (
      id,
      description,
      embedding vector_cosine_ops,
      category,
      rating,
      in_stock,
      created_at,
      last_updated_date,
      latest_available_time,
      metadata,
      weight_range
    );
    """);

If search_idx already exists, drop it first with DROP INDEX search_idx;.

Every indexed column is a field you can search, filter, sort, group, or aggregate. The key_field index option is deprecated and ignored. The description column is indexed for full-text search, embedding is indexed for vector search, and fields like rating, category, and created_at are available for filters, Top K queries, and facets. For production schemas, define this index in your migration system instead of running ad hoc SQL. The full Create an Index reference includes framework-native examples and tokenizer options. Next, run your first queries.

Getting Started

Next Steps

Create Your First Index