SQL-like query language and CLI for Qdrant vector search engine
# Install $ pip install qql-cli # Connect to Qdrant $ qql connect --url http://localhost:6333 # Insert a document — text is embedded automatically qql> INSERT INTO COLLECTION notes VALUES {'text': 'Qdrant is a vector database', 'year': 2024} ✓ Inserted 1 point [3f2e1a4b-...] # Search with a WHERE filter qql> SEARCH notes SIMILAR TO 'vector storage engines' LIMIT 5 WHERE year >= 2023 ✓ Found 1 result(s) # Hybrid dense+sparse search with cross-encoder reranking qql> SEARCH notes SIMILAR TO 'vector databases' LIMIT 5 USING HYBRID RERANK ✓ Found 1 result(s) (hybrid, reranked)
Installation, connecting to Qdrant, your first queries
Adding documents, batch inserts, payload types
Semantic search, point retrieval, pagination, hybrid search, reranking, recommendations
Full SQL-style filter operators — range, IN, MATCH, AND/OR/NOT
CREATE, DROP, quantize (scalar/binary/product), payload indexes
Run script files, export collections for backup/migration
Use QQL as a Python library without the CLI
Embedding models, config file, project structure, error guide