marqo-ai/marqo-ecommerce-embeddings

State-of-the-art embedding models fine-tuned for the ecommerce domain. +67% increase in evaluation metrics vs ViT-B-16-SigLIP.

21
/ 100
Experimental

Built on OpenCLIP with multimodal vision-language architecture, these models enable simultaneous text-to-image and image-to-text retrieval for product search. Two size variants (203M and 652M parameters) support deployment flexibility, with inference optimized for GPU acceleration (5-11ms per batch on A10G). Includes evaluation datasets (GoogleShopping-1m, AmazonProducts-3m) and integrates with both HuggingFace Transformers and OpenCLIP loaders for seamless adoption.

No commits in the last 6 months.

No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 5 / 25

How are scores calculated?

Stars

45

Forks

2

Language

Python

License

Last pushed

Nov 13, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/marqo-ai/marqo-ecommerce-embeddings"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.