Open-source search engine with vector storage and hybrid keyword-plus-semantic retrieval, self-hosted or run as a cloud service
Meilisearch is a search engine developed by Meili, a French company registered in Paris. The engine is open source and can be self-hosted; Meilisearch Cloud is the managed version of the same thing. Beyond typo-tolerant keyword search it stores embeddings and runs hybrid retrieval that blends lexical and semantic matching, which is what the RAG positioning rests on: it acts as the retrieval layer feeding an LLM, with context formatting and connectors for LangChain and LlamaIndex. The vendor cites sub-50ms queries across millions of documents. Cloud billing is usage or resource based and enterprise plans add SLAs, SSO and SOC 2.
The engine itself is open source and free to self-host. Meilisearch Cloud starts at $20/month with a 14-day trial and no card required; the cost estimator quoted $23/month for an XS instance and $30/month for 100K documents and 50K searches. Enterprise, with up to 99.999% uptime SLA, SSO SAML and SOC 2, is quoted.
Use tool ↗Meilisearch is the pragmatic middle between bolting pgvector onto Postgres and running an Elasticsearch cluster: hybrid retrieval works out of the box and the open-source licence means the cloud bill is a convenience, not a lock-in. The limit is scope — it is a retrieval layer, not a full RAG framework, and there is no permanently free cloud tier.
Watch out: The legal terms name Meili SAS in Paris, so cloud contracts are French-law agreements — worth checking against your data-residency requirements.
No reviews yet — be the first to review Meilisearch.
Reviews are tied to your EffectHub account — one review per tool, so the rating for Meilisearch reflects real users.
No questions yet — ask the first one about Meilisearch.
Questions about Meilisearch are tied to your EffectHub account, so answers can reach you and the section stays free of spam.