The Python Software Foundation has released Python 3.15.0 alpha 6, the sixth of eight planned alpha builds in the 3.15 series, offering developers an early preview of performance upgrades and long-awaited language features.
This pre-release introduces a substantial upgrade to the JIT compiler, delivering a 3–4% geometric mean performance boost on x86-64 Linux and a 7–8% speedup on AArch64 macOS compared to previous interpreters.
“Alpha releases are designed to make it easier to test the current state of new features and bug fixes, and to validate the release process itself,” said Hugo van Kemenade, Python release team member. “We encourage developers to help us shake out issues—but please keep this away from production environments.”
Key Changes in Python 3.15.0a6
The alpha introduces several PEPs that reshape how Python code is written and executed:
- PEP 799 – A new high-frequency, low-overhead statistical sampling profiler and dedicated profiling package.
- PEP 798 – Unpacking in comprehensions using
*and**. - PEP 686 – Python now uses UTF-8 as the default encoding for opening files, aligning with modern text processing.
- PEP 782 – A new PyBytesWriter C API for creating Python bytes objects more efficiently.
- PEP 728 – TypedDict with typed extra items, improving type safety for dictionaries.
Error messages have also been improved across the board, making debugging easier for both beginners and experienced developers.
Background: The Alpha Release Process
Python 3.15 is still in active development. The 3.15.0a6 release is the sixth alpha in a series of eight. During the alpha phase, features may be added or modified until the beta phase begins on 2026-05-05; further changes are possible until the release candidate phase on 2026-07-28.
This release is a developer preview only and is explicitly not recommended for production use. The next pre-release, 3.15.0a7, is scheduled for 2026-03-10.
For full details, see the official release page and the PEP 790 release schedule.
What This Means for Developers
The JIT improvements alone could significantly reduce runtime for compute-intensive workloads, making Python more competitive in performance-sensitive domains. The default UTF-8 encoding (PEP 686) simplifies cross‑platform file handling, while the new profiler (PEP 799) gives developers deeper insight into program performance without heavy overhead.
TypedDict enhancements (PEP 728) and unpacking in comprehensions (PEP 798) refine Python’s type system and expressiveness, though these features will remain experimental until the beta freeze.
Developers are urged to test their libraries and applications against this alpha and report bugs at github.com/python/cpython/issues.
Get Involved
The Python Software Foundation welcomes contributions—both code and financial support. “Thanks to all the volunteers who make Python development possible,” added Van Kemenade. “Consider supporting us through GitHub Sponsors or directly via the Python Software Foundation.”
For more resources, visit the online documentation.