Performance

Catch the import. Name the culprit. Keep threads free: Introducing gilguard On free-threaded Python, one import can quietly turn the GIL back on for good. gilguard turns that moment into an error that names the module, the package, and the line of your code that pulled it in.
Chameleon templates, modernized: chameleon-rs I rewrote Chameleon in Rust to make Python web apps faster. chameleon-rs runs your existing templates unchanged and adds async rendering, Markdown, and free-threaded Python support.
Cutting Python Web App Memory Over 31% I reduced Python web app memory by 3.2 GB using async workers, import isolation, the Raw+DC database pattern, and disk caching. Detailed before and after numbers for each technique.
Raw+DC Database Pattern: A Retrospective After migrating three production Python web apps to the Raw+DC database pattern, I share real-world performance benchmarks: nearly 2x requests per second, 18% less memory, native async support, and freedom from unmaintained ODM libraries.
Replacing Flask with Robyn wasn't worth it I converted Python Bytes from Quart/Flask to the Rust-backed Robyn framework and benchmarked it with Locust. The result: no meaningful speed or memory improvement. Here's why framework maturity still wins.
Fire and forget (or never) with Python’s asyncio Python's asyncio.create_task() can silently garbage collect fire-and-forget tasks in 3.12+. Learn the background_tasks set pattern to fix it.
Python Numbers Every Programmer Should Know A cheat sheet of real-world timing and memory numbers to guide performance-sensitive decisions.
We Must Replace uWSGI With Something Else uWSGI is in minimal maintenance mode, now what?
Let's go easy on PyPI, OK? Why do we pound PyPI when we have caching?
Docker images using uv's python uv has added a new way to build Python-based containers.
Black Friday: A Lesson in Python Performance Black Friday almost melted our servers
20% Faster Python with a Single GC Tweak Your Python GC settings are wrong and they are hurting your performance.