Trieve is an all-in-one platform designed for developers to integrate search, recommendations, Retrieval-Augmented Generation (RAG), and analytics into their applications. It offers a comprehensive API and SDKs, with a focus on self-hosting and flexibility in model integration.

How It Works

Trieve employs a hybrid search approach, combining semantic vector search using OpenAI or Jina embeddings with Qdrant, and typo-tolerant neural sparse-vector search powered by naver/efficient-splade-VI-BT-large-query. It also supports cross-encoder re-ranking with models like BAAI/bge-reranker-large for enhanced relevance. Features like sub-sentence highlighting, recency biasing, and tunable merchandising are included to optimize search UX.

Quick Start & Requirements

• Installation: Self-hosting guides are available for AWS, GCP, Kubernetes, and Docker Compose. Local development requires Rust, Node.js (via NVM), and Yarn.
• Prerequisites: OpenAI API key (or other LLM provider via OpenRouter), curl, gcc, g++, make, pkg-config, python3, python3-pip, libpq-dev, libssl-dev, openssl, postgresql-libs.
• Setup: Local development involves cloning the repo, installing dependencies, setting environment variables (including LLM API keys), building Rust components, and running Docker services.
• Documentation: API Reference + Docs

Highlighted Details

• Supports semantic, full-text, and hybrid search with cross-encoder re-ranking.
• Offers RAG API routes with topic-based memory management or custom context selection.
• Enables bringing your own embedding, SPLADE, cross-encoder, and LLM models.
• Includes features like sub-sentence highlighting, grouping by file, and filtering.

Maintenance & Community

• Active development with a small, hands-on team.
• Community channels available via Discord and Matrix.
• Professional services are offered.

Licensing & Compatibility

• The project appears to be open-source, but a specific license is not explicitly stated in the README. Compatibility for commercial use or closed-source linking would require clarification on licensing terms.

Limitations & Caveats

• The README does not explicitly state the open-source license, which is critical for commercial adoption. Local setup instructions are extensive and may require significant time and system configuration.