Python-Performance-Benchmark

🧐 Comparing Python 3.10, 3.13 and 3.13(Free-threaded Mode) processing speed

Note

• Benchmark code performs factorial calculations

🖥️ Environment

OS	CPU	RAM	Python
Ubuntu 22.04	Intel(R) Core(TM) i5-8500B CPU @ 3.00GHz	8GB	3.10.16, 3.13.1t

🏆 Performance

Note

• GIL deactivation of Python 3.13 seems to be less optimized than pypy or cython yet

✏️ Prepare

1. Create Python 3.10, Python 3.13 environments with Anaconda and pyenv

   # Python 3.10
   conda create -n bechpy3.10 python=3.10
   
   # Python 3.13
   pyenv install 3.13.1t

Tip

• Python 3.13.1t has GIL disabled by default and can be turned on through arguments

📚 Usage

⭐ Python 3.13(Free-threaded Mode)

Important

• Please make sure it's a Python 3.13 environment

1. Check GIL activity

python GIL_check.py

Tip

• If the output is False, GIL is disabled

2. Run benchmark

python bench.py

⭐ Python 3.13

Important

• Please make sure it's a Python 3.13 environment

1. Run benchmark

python -X gil=1 bench.py

Tip

• -X gil=1 argument activates GIL
• In Python 3.13.1t, GIL is inactive by default and can be specified as gil=0

⭐ Python 3.10

Important

• Please make sure it's a Python 3.10 environment

1. Run benchmark

python bench.py