devflowinc/trieve

Health v2 · maintenance only

Adoption not part of the score

2717 stars · 252 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Trieve is an all-in-one API platform for semantic vector search, typo-tolerant full-text and neural search, recommendations, RAG, and analytics. It is built in Rust with Qdrant and PostgreSQL, and can be used via hosted API or self-hosted in a VPC with Docker Compose or Kubernetes.

Use cases

• add semantic search to my app
• build a RAG chatbot over my documents
• search pdfs with embeddings
• recommendation engine via api
• self-host a vector search service
• typo-tolerant full-text search
• analytics for search queries

When to choose

• you want search, recommendations, and RAG from a single API
• you need self-hosting in your own VPC or on-prem
• you want hybrid dense vector and SPLADE full-text search out of the box

When to avoid

• you only need a lightweight embedded vector store inside your app
• you want a fully managed service from a large cloud vendor
• you need a simple keyword-only search without embeddings infrastructure

Facets

service · maturity active

search-engine vector-database rag llm-inference analytics api-framework artificial-intelligence large-language-models apis self-hosted self-hosted cloud semantic-search recommendations qdrant splade hybrid-search mcp-server rust api-platform embedding search retrieval-augmented-generation docker kubernetes web-server

Member repositories

For agents

markdown · JSON · MCP: product_card(name="devflowinc/trieve")