1 of 13

Magento �AI Search �by Ceymox

www.ceymox.com

1

2 of 13

AI-Driven Search Experience

Uses Google Gemini Flash Model for advanced AI search, delivering more relevant and personalized search results based on product details.

2

3 of 13

Search Widget

A customizable search widget that can be placed anywhere on the store’s frontend.

Configurable minimum score for search result precision

Flexible Configuration

Supports manual vector creation via the Magento console.

Product Vector Creation

Real-time product suggestions with sorting and filtering options on the search results page.

Ability to define the number of search suggestions visible under the search result widget.

Automates vector creation using the Magento queue system for large datasets.

3

4 of 13

Synced Product Vector Status

A backend grid to track product vector creation and sync status

Full support for Elasticsearch 7 and OpenSearch 2.17.1 for flexible, high-performance searches.

Elasticsearch and OpenSearch Support

Option to enable cron jobs for syncing product vector data automatically, which is useful for high-volume stores.

Cron Job for Vector Syncing

4

5 of 13

Top Features

These features make the extension a cutting-edge solution for Magento stores aiming to improve search functionality and user experience.

1

AI-Powered Search:

Leverages advanced Google Gemini AI models to deliver highly accurate, intelligent search results, enhancing user experience and product discovery.

2

Natural Language Processing (NLP) Search:

Enables users to search in everyday language, making the search more intuitive and efficient, understanding complex queries with ease.

3

Machine Learning Enabled Search:

Uses Machine Learning models to continuously improve the relevance and personalization of search results based on customer behavior and product data.

4

Smart Search Solution:

Provides real-time, dynamic suggestions based on user input, improving conversion rates with faster and more accurate product recommendations.

5

6 of 13

5

Vector Search Technology:

Utilizes vector search algorithms powered by Elasticsearch and OpenSearch, ensuring lightning-fast, precise search results even for large catalogs.

6

Customizable Search Widget:

Fully customizable search interface that can be placed anywhere on the site, providing advanced filters and sorting options to enhance the user journey.

7

Elevate User Experience:

Transform your store’s search functionality into a seamless, intelligent experience. With natural language search, smart product suggestions, and real-time filters, users will love how easy it is to browse your catalog.

8

Seamless Integration with Magento:

The extension seamlessly integrates with Magento 2.4.x, offering easy installation, configuration, and updates. Our comprehensive documentation makes it simple to set up and customize for your unique needs.

6

7 of 13

About the Ceymox AI Module

Extension Info

This module was developed to perform vector search using elastic and open search engines. The product information is vectorized using the Google Gemini Text Embeddings 004 model and the search is done by the Google Gemini Gemini-1.5-flash model. The search widget can be placed anywhere you need and perform search actions. A corresponding search result page is also available with sort and filter options.

Module name → Ceymox_AIProductSuggestion

Module version → 1.0.0

Magento Tested version: 2.4.x

Tested PHP version: 8.1,8.2

Elasticsearch version 7 support

OpenSearch version 2.17.1 support

7

8 of 13

Getting the Top Result

To get proper results with a query we must add maximum keywords related to the product on its description for a better search experience

1

Example

If you have a product like a T-shirt you can add details like. material of the product, suitable age group, color, occasion, suitable climate, gender, gift type or not, size, brand,.....

2

Example

If you are running a grocery store and you have products like onion you can add its details like, it is a vegetable, used for making recipes, used recipe examples…

So while specifying all the details about the product user can easily fetch the right product with his query. The resulting query is completely based on what information you added to the product information.

8

9 of 13

Creating Vectors

You can create vectors in two ways

You can run below mentioned command to create vectors this is recommended for smaller amounts of data�php bin/magento ceymox:generate_embeddings

Using console Command

You can create vectors using the Magento queue system. It is useful when you have a larger number of products. Default the batch size is 20000 if you have a higher amount of data you must increase the batch size from the count specified in the env.php file

Using the Magento Queue system

1

2

9

10 of 13

Gemini API Key Generation

Before starting with this extension we need to create a Google Gemini API key for that follow the below steps

Navigate to https://ai.google.dev/ → Get the API Key in Google AI Studio

Click on the Create API key button (if you are not logged in you must be logged in before creating the API key)

10

11 of 13

Easy Installation & Customization

No complex setup is needed! Our extension is designed for seamless integration with your Magento store. With just a few steps, you can configure the AI search module, connect it with Google Gemini AI, and start providing a better search experience to your customers.

11

12 of 13

Start Transforming Your Store Today!

Ready to revolutionize your eCommerce experience? Empower your customers with intelligent, AI-driven search functionality that delivers better results, faster.

Get started with the Magento 2 AI Search Extension by Ceymox today and see the difference AI can make in your sales and customer satisfaction.

12

13 of 13

Download Extension

https://shop.ceymox.com/

1, Go to Shop.Ceymox.com

2, Select AI Search Extension

3, Add to Cart and Checkout

Please scan the QR code below to download the extension, or you can browse to shop.ceymox.com.

13