For compiling from source (required for free-threading or JIT features):
: This remains an optional, experimental binary flag. You must download or build the dedicated free-threaded version (e.g., configuring with --disable-gil or choosing the experimental option in macOS/Windows installers). 2. Experimental Copy-and-Patch JIT Compiler (PEP 744)
The JIT is not enabled by default. You must compile CPython from source with the --enable-experimental-jit flag. Even then, it only compiles relatively small parts of the interpreter's bytecode dispatch loop.
This release lays the foundational infrastructure for a native . Utilizing a "copy-on-write" template engine technique, the experimental Tier-2 JIT compiles specific internal bytecodes directly into machine code at runtime.
: Users can edit blocks of multi-line functions interactively without breaking the shell flow. python 313 release notes verified
The internal handling of scope and memory has been refined to increase stability and performance.
It is considered foundational work that will be refined in Python 3.14 and beyond. 4. Improved locals() and Memory Management
: This is one of the most significant changes in Python's history. It allows users to disable the Global Interpreter Lock (GIL)
One universally available improvement requires no special compilation: the interactive interpreter (REPL) has been completely rewritten based on PyPy's implementation. The new REPL supports multi-line editing with seamless block handling, colorized syntax highlighting, and persistent command history across sessions. Exception tracebacks are now colorized by default, with different colors distinguishing error types, file paths, and line numbers. For compiling from source (required for free-threading or
As of April 2026, Python 3.13 has moved into its maintenance phase. The latest available stable maintenance release is , which includes approximately 240 bug fixes and security improvements.
Python 3.13 is foundational , not flashy. The verified changes point toward a multicore future — but we're not there yet. Upgrade, experiment, report bugs.
: You can now type exit , quit , or clear directly without calling them as functions (e.g., exit() is no longer required). 3. Language & Standard Library Updates
The performance gains, when applicable, can be substantial. Preliminary benchmarks show that free-threading significantly benefits computationally intensive parallel workloads, with speedups scaling reasonably with core counts. Experimental Copy-and-Patch JIT Compiler (PEP 744) The JIT
For the first time, developers can run CPython with the .
Deprecated multimedia format utilities removed from core memory. SQLite Engine Defaults for dbm
Tier 2 optimizer improvements → 5–15% faster for some pure-Python loops.