About

Marqo's specialized embedding models for ecommerce are designed to help online retailers improve their search, retrieval, and recommendation tasks. These models, specifically Marqo-Ecommerce-B and Marqo-Ecommerce-L, are optimized for ecommerce and offer enhanced performance in real-world search scenarios, making them ideal for businesses looking to enhance customer experience and increase revenue.

Details

The key features of Marqo's ecommerce embedding models include:

Multimodal product embeddings: Generate embeddings from images and text to enable accurate search and retrieval
Ecommerce-specific optimization: Trained on 100s of millions of samples from ~50 million unique products across 20,000 Amazon ASIN categories
High-performance results: Outperform existing state-of-the-art solutions by up to 88% in certain tasks
Easy integration: Available on Hugging Face and Marqo Cloud, with easy-to-use APIs and documentation
Customizable: Can be fine-tuned on your own product catalogs and user behavior using Marqtune

Marqo's ecommerce embedding models, Marqo-Ecommerce-B and Marqo-Ecommerce-L, offer a significant improvement over existing solutions for search, retrieval, and recommendation tasks in