Typesense Search PRO

Typesense Search Pro replaces Magento's default search engine with a fully native Typesense integration, giving merchants a choice most search extensions don't: run on Typesense Cloud for zero-maintenance hosting, or self-host your own cluster for full data control - complete with health checks, metrics, backup snapshots, and database compaction tools built into the Magento admin.
Where this extension goes beyond standard search is AI. Shoppers can search the way they'd ask a person - "waterproof jacket for hiking under $200" - and the module's natural language engine (GPT-4o, Gemini, Vertex AI, Cloudflare Workers AI, or a self-hosted vLLM instance) translates that into structured Typesense filters automatically. Pair that with semantic and vector search using S-BERT, GTE, E-5, or OpenAI embeddings, and results surface by intent and meaning, not just keyword overlap.
Underneath the AI layer sits a complete, production-grade search stack: instant autocomplete, faceted filtering and sorting, configurable synonym and typo-tolerance rules, queue- and cron-based incremental indexing so catalog changes reach the index without full reindexes, a GraphQL API for headless and PWA storefronts, and search curation tools so merchants can pin or hide specific products for any given query.

Features:-

1. Native Typesense Engine
Fully replaces Magento's OpenSearch/Elasticsearch search engine at the core — not a supplementary widget layered on top of default search.

2. Cloud or Self-Hosted Deployment
Connect to Typesense Cloud for managed hosting, or point the module at your own self-hosted Typesense cluster. Self-hosted deployments get in-admin cluster management: health checks, live metrics, backup snapshots, and database compaction.

3. AI-Powered Natural Language Search
Shoppers type queries in plain language, and the module converts them into structured Typesense filters using a configurable LLM backend — GPT-4o, Gemini, Vertex AI, Cloudflare Workers AI, or a self-hosted vLLM model.

4. Semantic & Vector Search
Matches by meaning, not just keyword overlap, using embedding models including S-BERT, GTE, E-5, and OpenAI.

5. Instant Autocomplete
Live suggestions across products, categories, CMS pages, and query suggestions as the shopper types.

6. Faceted Search & Filtering
Full faceted result pages with attribute filters and configurable sort options.

7. Synonym & Typo Tolerance Management
Configure multi-way and one-way synonyms, plus typo-tolerance thresholds, directly from the admin.

8. Queue & Cron-Based Incremental Indexing
Catalog changes sync to Typesense via async queue consumers or scheduled cron, avoiding full reindex overhead.

9. GraphQL API for Headless/PWA Storefronts
Exposes search, autocomplete, and filtering through GraphQL for PWA Studio, Hyvä, and other headless frontends.

10. Search Curation & Merchandising
Pin or hide specific products for any given search query to control result ranking manually.