The Best of Keywords and Semantics
GNO combines BM25 full-text search with vector similarity search using reciprocal rank fusion. Get accurate results whether you search for exact terms or conceptual matches.
Key Benefits
- Exact keyword matching with BM25
- Semantic similarity with vector search
- Reciprocal rank fusion combines both
- Configurable weights for tuning
Example Commands
gno search 'exact terms'
gno vsearch 'conceptual query'
gno query 'best of both'
Get Started
Ready to try Hybrid Search?
How It Works
GNO’s hybrid search combines two powerful search paradigms:
BM25 Full-Text Search
Traditional keyword matching that excels at finding exact terms. When you search for “authentication JWT”, BM25 finds documents containing those exact words.
gno search "authentication JWT"
Vector Similarity Search
Semantic search using embeddings. When you search for “how to protect my app”, vector search finds documents about security, authentication, and access control - even without those exact words.
gno vsearch "how to protect my app"
Reciprocal Rank Fusion
The query command combines both methods using RRF, a proven algorithm that merges ranked lists. Documents that score well in both methods rise to the top.
gno query "authentication best practices"