Do AI agents pick Orama?
How often agents choose Orama when a developer needs search — measured across Claude, GPT, Gemini.
95% CI 0%–3%
When developers ask an AI agent for search, Orama is picked 0% of the time — ranking #5 of 5 measured across Claude, GPT, Gemini.
1.0M package downloads / week
0 docs in FineWeb · 5 URLs in Common Crawl
3.4M installs / month ▲ 21% MoM
Get the full report
The per-model and per-surface breakdown for Orama — where it wins, where it loses, and to whom — plus an alert when the Pick Rate moves.
About Orama
An open-source search engine that runs full-text, vector, and hybrid search entirely in JavaScript — no external server required.
Orama is a JavaScript search engine and RAG pipeline that runs in the browser, on the server (Node.js, Deno, Bun), or at the edge. It supports full-text search with BM25 ranking, vector search using embeddings, hybrid search combining both, geosearch, facets, typo tolerance, filters, stemming across 30 languages, and GenAI chat sessions. It is aimed at developers who want capable search without standing up a dedicated search service.
Where to start
Install the @orama/orama package via npm or import it directly from a CDN, then create a database with a typed schema, insert documents, and call the search function. The official documentation covers all search modes, plugins, and configuration options.
The rest of the Search ranking
| 1 | Meilisearch | 17% |
| 2 | Algolia | 14% |
| 3 | Elasticsearch | 8% |
| 4 | Typesense | 4% |