Key features of Trieve include:
- Managed Embedding Models: Bring your own embedding model or use one of the available open source defaults.
- SPLADE Full-text Neural Search: Utilize SPLADE, a state-of-the-art retrieval model for full-text search.
- Semantic Vector Search: Trieve supports semantic vector search out of the box.
- Hybrid Search: Combine full-text search and semantic vector search with cross-encoder re-ranker models.
- Merchandising Relevance Tuning: Boost search results based on sales or popularity.
- Date Recency Biasing: Ensure the most recent results are shown first with date recency biasing.
- Open Source Models: Both default embedding model and LLM are open source.
- Batteries Included: Comprehensive API surface with thorough documentation.
- Host Yourself: Host sensitive data with terraform templates and no external dependencies.
- Search, Recommend, or Generate: Easily integrate search and RAG API calls into your product.