Release and Version History¶

x.y.z (Backlog)¶

Features and Improvements

Minor Improvements

Bugfixes

Miscellaneous

First Release

Features and Improvements

S3 Vector Bucket Management: Create, delete, and list vector buckets with automatic conflict handling
Vector Index Operations: Full CRUD operations for vector indexes including creation, retrieval, and deletion
Type-Safe Vector Models: Pydantic-based vector classes with automatic data validation and metadata support
ORM-Style Query Builder: SQLAlchemy-inspired metadata filtering with support for all comparison operators (eq, ne, gt, gte, lt, lte, in_, nin, exists)
Complex Query Combinations: Logical operators (AND, OR) for building sophisticated metadata filters
Vector Similarity Search: High-performance similarity search with configurable distance metrics and top-K results
Pagination Support: Built-in pagination for listing vectors and indexes with configurable page sizes
Segmented Processing: Parallel processing support through vector segmentation for large datasets
Bulk Operations: Efficient batch vector insertion and deletion operations
Type Preservation: Generic type system that preserves Vector subclass types throughout query operations

Vector Management

put_vectors(): Store multiple vectors with metadata in batch operations
query_vectors(): Similarity search with optional metadata filtering and distance calculation
list_vectors(): Paginated vector listing with optional data and metadata return
delete_vectors(): Selective vector deletion by keys
delete_all_vectors(): Convenient method to clear all vectors from an index
as_vector_objects(): Convert AWS responses back to strongly-typed Vector instances

Metadata System

BaseMetadata: Declarative metadata models with inheritance support
MetaKey: Fluent query builder for all AWS S3 Vectors filtering operators
Automatic Field Registration: Metaclass-driven field discovery and registration
Query Expression Building: Natural Python syntax for complex filter construction

Developer Experience