Summa

This project is archived. Summa's successor is Hermes — a high-performance, embeddable full-text search engine with BM25 ranking, WAND optimization, sparse vector search, gRPC API, and WebAssembly support.

Summa was a full-text IPFS-friendly search engine that could be launched on both large servers and inside your browser.

Successor: Hermes

Hermes carries forward and extends Summa's vision with:

• Async search engine written in Rust with BM25 ranking and WAND/MaxScore optimizations
• Sparse vector search (SPLADE, learned sparse retrieval) with BMP index
• gRPC server for remote search operations
• WebAssembly bindings for browsers
• Python client library
• CLI for index management and data processing pipelines