sqlalchemy-model-generator

To generate SQLAlchemy models with pgvector support, provide the table name and field specifications. The tool scaffolds model classes with UUID primary keys, vector embedding columns, and BM25 full-text search integration consistent with existing conventions.

Adding vector embeddings to a SQLAlchemy model is handled by creating pgvector columns with dimensions controlled by application settings. This ensures vector search capabilities are integrated while maintaining alignment with existing database schemas and retrieval scaffolds.

BM25 full-text search with SQLAlchemy uses TSVector columns and appropriate indexes to enable fast text lookups. The model generator creates these indexing scaffolds automatically, allowing ML-assisted apps to perform efficient retrieval alongside vector similarity searches.

Yes, the generated SQLAlchemy model strictly enforces UUID primary keys for all new entities. This standardization ensures consistency across your database tables and aligns with existing model conventions for reliable data lineage and integration.

You can update initialization and retrieval scaffolds like init_db.py and retrieval services automatically when adding new tables. The generator reads and references existing models to seamlessly integrate new entities while maintaining structural consistency.

Scaffolding SQLAlchemy models for vector search is limited to PostgreSQL databases using pgvector. It requires existing model conventions to reference for consistency and focuses on generating UUID-based entities, so it may not suit legacy schemas lacking these structural foundations.