
All search types
Search playground
Search 82,000 apparel products with full-text, vector, hybrid and image search or ask AI, and see the SQL behind every result.
- Full-text, vector, and hybrid search
- Search by photo
- SQL and API for every query
Live, public examples
Try real search interfaces powered by Manticore Search. Each demo focuses on a concrete search pattern and lets you inspect the behavior in a working application.
Try the public datasets and working interfaces below to explore different retrieval and discovery patterns.

All search types
Search 82,000 apparel products with full-text, vector, hybrid and image search or ask AI, and see the SQL behind every result.

Guided navigation
Filter a realistic storefront and compare strict, auto, and max modes while products and facet counts stay consistent.

Product discovery
Search a board-game catalog with autocomplete, filters, facets, typo tolerance, semantic retrieval, and similar-item recommendations.

GitHub issues
Run keyword and operator searches over Manticore Search issues and pull requests.

Meaning-based retrieval
Find related GitHub issues by meaning rather than relying only on exact words.

Search as you type
Type a prefix and see live query suggestions generated over the GitHub issue dataset.

Typo tolerance
See how a misspelled query can still recover relevant GitHub issue results.

Multimodal retrieval
Search an image collection with visual similarity powered by vector embeddings.

Natural-language shopping
Ask natural shopping questions over 82,000 apparel items and inspect the resulting product recommendations.
Location-based search
Manticore stores latitude and longitude as document attributes and filters by distance right in the index. Drag the pin, pull the radius, and watch the nearest products update.
Build your own
Install Manticore Search, load your data, and use SQL or JSON APIs to build the search experience your product needs.