The third beta release of the SQLAlchemy 2.1 series is now available.

Release 2.1.0b3 brings a focused set of new features, performance improvements, and bug fixes. With no major regressions reported against 2.1.0b2, we are hoping this release will be the last beta before a final 2.1.0 release, targeting end of summer 2026.

Highlights of this release include:

• ORM Loading Performance Improvements - ORM result row fetching now processes rows as plain tuples rather than constructing Row objects, since ORM loaders use position-based access and do not require the Row interface. Additional optimizations were made to selectinload and subqueryload result handling. Benchmarks show a 3-16% improvement in ORM entity load times depending on query shape. Thanks to Oliver Parker for this work.
• selectinload omit_join for Many-to-Many - The selectinload loader strategy now applies the omit_join optimization automatically for many-to-many non-self-referential relationships, selecting from the secondary (association) table directly rather than joining back to the parent entity, reducing the number of JOINs in the emitted SELECT.
• SQLite JSONB Type - A new JSONB type supports SQLite's binary JSON storage format, available as of SQLite version 3.45.0. Values are stored and retrieved using the jsonb() / json() SQL functions while Python-side behavior remains identical to JSON. Thanks to Shamil Abdulaev.
• selectinload chunksize Parameter - A new chunksize parameter on selectinload allows configuring the number of primary keys sent per IN clause when loading relationships, giving applications control over query size for very large result sets. Thanks to bekapono.
• Delete.using() for Multi-Table DELETE - A new Delete.using() method allows explicit FROM expressions to be rendered in backend-specific multiple-table DELETE forms, including MySQL/MariaDB DELETE .. USING. Thanks to cjc0013.
• SQL Server Bulk Reflection - The SQL Server dialect now implements native multi-table reflection methods, issuing a single bulk query per object type against the sys.* catalog views rather than a per-table round trip loop, matching the pattern already used by the PostgreSQL and Oracle dialects. Thanks to Gaurav Sharma.
• Session execution_options for Unit of Work Core SQL - Session-level execution_options now take effect for Core-level SQL emitted by unit of work operations, in addition to ORM statement executions. This allows options such as schema_translate_map to be applied to a Session globally.

SQLAlchemy 2.1 includes over 90 individual documented changes across all betas, so be sure to check out not just the What's New in SQLAlchemy 2.1? migration guide, but the detailed changelog at Changelog as well.

SQLAlchemy 2.1.0b3 by naming convention is a beta release, so only installs with pip if the --pre flag is included. Links to pypi for SQLAlchemy 2.1.0b3 are available on the Download Page.

SQLAlchemy 2.0.51 is now available.

Links to the detailed changelog for 2.0.51 is at Changelog.

SQLAlchemy 2.0.51 is available on the Download Page.

SQLAlchemy 2.0.50 is now available.

Links to the detailed changelog for 2.0.50 is at Changelog.

SQLAlchemy 2.0.50 is available on the Download Page.

The second beta release of the SQLAlchemy 2.1 series is now available.

Release 2.1.0b2 adds new support for the mssql-python DBAPI, a driver for Microsoft SQL Server produced directly by Microsoft. A few other features and bug fixes are also introduced.

Depending on feedback for release 2.1.0b2, the next release may be 2.1.0 final, unless further issues that can benefit from beta testing are added.

SQLAlchemy 2.1 includes over 90 individual documented changes, so be sure to check out not just the What's New in SQLAlchemy 2.1? migration guide, but the detailed changelog at Changelog as well.

SQLAlchemy 2.1.0b2 by naming convention is a beta release, so only installs with pip if the --pre flag is included. Links to pypi for SQLAlchemy 2.1.0b2 are available on the Download Page.

SQLAlchemy 2.0.49 is now available.

Links to the detailed changelog for 2.0.49 is at Changelog.

SQLAlchemy 2.0.49 is available on the Download Page.

SQLAlchemy 2.0.48 is now available.

Release 2.0.48 fixes a single potential concurrency issue which could impact custom "on connect" routines, particularly those used by the SQL Server dialect.

Links to the detailed changelog for 2.0.48 is at Changelog.

SQLAlchemy 2.0.48 is available on the Download Page.

SQLAlchemy 2.0.47 is now available.

Links to the detailed changelog for 2.0.47 is at Changelog.

SQLAlchemy 2.0.47 is available on the Download Page.

SQLAlchemy 2.0.46 is now available.

Release 2.0.46 includes typing improvements for ORM mapped classes and aliased entities in result row mappings, several PostgreSQL dialect fixes including corrections to JSONB operators and improved unicode support in foreign key reflection, and enhancements to the MariaDB, SQLite, and MSSQL dialects.

Links to the detailed changelog for 2.0.46 is at Changelog.

SQLAlchemy 2.0.46 is available on the Download Page.

The first beta release of the SQLAlchemy 2.1 series is now available.

Nearly two years in development, the 2.1 series builds upon the new Python typing-oriented architecture of SQLAlchemy 2.0, introducing a range of enhancements and new capabilities across both Core and ORM components, with an emphasis on newer Python idioms such as dataclasses, improved pep-484 typing, template strings, and continued accommodations for Python free threading.

Key improvements in SQLAlchemy 2.1 include:

• Python 3.14 Template String Support - A new tstring() construct works similarly to the text() construct, but works with the new t'' syntax added in Python 3.14+, allowing for automatic generation of bound parameters from a fully inline SQL string expression.
• CREATE VIEW and CREATE TABLE AS Support - New DDL constructs enable creation of views and table-as-select operations
• pep-646 typing for SQL SELECT statements, result sets, rows - The Tuple[] seen in type annotations for selectable statements and results has been replaced by the use of pep-646 Unpack directly, representing heterogeneous rows and supporting constructs directly without the need to manually unpack results from a tuple for typing of individual unpacked values to work.
• Refined Dataclass Mapping - ORM mapped dataclasses no longer populate implicit defaults in __dict__, resolving workflow issues where foreign key values would incorrectly override relationship attributes
• Default driver updates for PostgreSQL, Oracle - The default driver for postgresql:// URLs is now psycopg, replacing psycopg2, and the default driver for oracle:// URLs is now oracledb, replacing cx_oracle. This is the first release in SQLAlchemy's history to revise these default drivers.
• Support for monotonic server side functions such as PostgreSQL's uuidv7() in batched INSERTs - SQLAlchemy's "insertmanyvalues" feature introduced in version 2.0 allowed dramatic improvements for INSERT statements as used by the ORM, when primary key values needed to be fetched from newly inserted rows without executing row-at-a-time. Release 2.1 adds support for the use of monotonically increasing server side functions to be integrated within this system, with PG's uuidv7() construct as the current example.
• Greenlet is a fully optional dependency - SQLAlchemy's asyncio support in version 2.1 remains dependent on greenlet to switch between asyncio and synchronous internal logic. However, as greenlet continues to have sporadic compatibility issues when new Python versions are released, it no longer installs automatically and must be installed separately, typically using the sqlalchemy[asyncio] target. This change also supports Python free threading, as greenlet as of this writing is not yet compatible with free-threaded Python builds

SQLAlchemy 2.1 includes over 90 individual documented changes, so be sure to check out not just the What's New in SQLAlchemy 2.1? migration guide, but the detailed changelog at Changelog as well.

SQLAlchemy 2.1.0b1 by naming convention is a beta release, so only installs with pip if the --pre flag is included. Links to pypi for SQLAlchemy 2.1.0b1 are available on the Download Page.

SQLAlchemy 2.0.45 is now available.

Links to the detailed changelog for 2.0.45 is at Changelog.

SQLAlchemy 2.0.45 is available on the Download Page.