cortexdblisted

# CortexDB CortexDB is a lightweight SQLite-based vector database for Go AI projects. It supports vector similarity search, hybrid search (vector + keyword), full-text search (FTS5/BM25), knowledge graphs, and memory APIs. ## Core Concepts ### Vector Search Store and search embeddings (vector representations of data). CortexDB supports multiple index types (HNSW, IVF, Flat) and similarity functions (cosine, dot product, euclidean). ### Hybrid Search Combines vector search with FTS5 keyword search using Reciprocal Rank Fusion (RRF). ### Lexical Fallback (v2.13.0+) When no embedder is configured, `SearchText` and `HybridSearchText` automatically fall back to FTS5/BM25 full-text search. ### LLM-Assisted Retrieval (v2.13.0+) Query expansion using LLM-generated keywords and alternate query phrasings for improved FTS5 recall. ### External Vector Support (v2.13.0+) APIs for ingesting pre-computed vectors from external pipelines without requiring an embedder. ## Installation ```go import "github.com/liliang-cn/cortexdb/v2" ``` ## Quick Start ```go import "github.com/liliang-cn/cortexdb/v2/pkg/cortexdb" // Open database db, err := cortexdb.Open(cortexdb.DefaultConfig("/path/to/db")) if err != nil { log.Fatal(err) } defer db.Close() // Quick operations q := db.Quick() // Add vector with auto ID id, err := q.Add(ctx, []float32{0.1, 0.2, 0.3}, "document content") // Search results, err := q.Search(ctx, []float32{0.1, 0.2, 0.3}, 10) // Text search (requires embedder) r