Case Study: Hugging Face improves discovery across 2M+ AI assets with Meilisearch

Hugging Face powers discovery of 2M AI assets with Meilisearch

Hugging Face, an open-source provider of machine learning technologies, faced a challenge in ensuring the discoverability of over 2 million AI models, datasets, and demos on its Hub platform. Their existing simple filtering and keyword search solution was insufficient for the growing demand for a more flexible, typo-tolerant full-text search that could leverage valuable metadata. To address this, they turned to vendor Meilisearch.

The solution implemented was Meilisearch Cloud, chosen for its customizability, ease of use, and quality of support. Meilisearch now powers full-text search across the entire content of model cards, providing users with higher default relevancy. The measurable impact is that Meilisearch currently enables the discovery of 2.2 million model cards, 500,000 datasets, and 60,000 demos, actively supporting Hugging Face's mission of AI democratization.

Hugging Face