From Async/Await to Virtual Threads
Armin Ronacher revisits his earlier critique of async/await in Python, arguing for a simpler and more composable concurrency model built around structured, virtual threads that avoid colored functions and simplify cancellation and context passing. He proposes rethinking Python’s concurrency by embracing thread-first APIs with virtual thread groups, which can yield, cancel, and propagate ...
https://lucumr.pocoo.org/2025/7/26/virtual-threads/
Armin Ronacher revisits his earlier critique of async/await in Python, arguing for a simpler and more composable concurrency model built around structured, virtual threads that avoid colored functions and simplify cancellation and context passing. He proposes rethinking Python’s concurrency by embracing thread-first APIs with virtual thread groups, which can yield, cancel, and propagate ...
https://lucumr.pocoo.org/2025/7/26/virtual-threads/
Armin Ronacher's Thoughts and Writings
From Async/Await to Virtual Threads
A follow-up to how I wish async would work.
QwenLM / Qwen3
Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.
https://github.com/QwenLM/Qwen3
Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.
https://github.com/QwenLM/Qwen3
GitHub
GitHub - QwenLM/Qwen3: Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.
Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud. - QwenLM/Qwen3
I built an offline, open‑source desktop Pixel Art Editor in Python
https://github.com/danterolle/tilf
https://github.com/danterolle/tilf
GitHub
GitHub - danterolle/tilf: Tilf (Tiny Elf) is a simple yet powerful pixel art editor built with PySide6. It’s designed for creating…
Tilf (Tiny Elf) is a simple yet powerful pixel art editor built with PySide6. It’s designed for creating sprites, icons, and small 2D assets with essential drawing tools, live preview, undo/redo, a...
PyPI: Preventing ZIP parser confusion attacks on Python package installers
https://blog.pypi.org/posts/2025-08-07-wheel-archive-confusion-attacks/
https://blog.pypi.org/posts/2025-08-07-wheel-archive-confusion-attacks/
blog.pypi.org
Preventing ZIP parser confusion attacks on Python package installers - The Python Package Index Blog
PyPI will begin warning and will later reject wheels that contain differentiable ZIP features or incorrect RECORD files.
Forget metaclasses; Python’s `__init_subclass__` is all you really need
https://www.reddit.com/r/Python/comments/1mevs3i/forget_metaclasses_pythons_init_subclass_is_all/
https://www.reddit.com/r/Python/comments/1mevs3i/forget_metaclasses_pythons_init_subclass_is_all/
Reddit
From the Python community on Reddit
Explore this post and more from the Python community
HRT's Python Fork: Leveraging PEP 690 for Faster Imports
https://www.hudsonrivertrading.com/hrtbeat/inside-hrts-python-fork/
https://www.hudsonrivertrading.com/hrtbeat/inside-hrts-python-fork/
Hudson River Trading
Inside HRT’s Python Fork: Leveraging PEP 690 for Faster Imports
At HRT, we’ve found that centralizing our codebase facilitates cross-team collaboration and rapid deployment of new projects. Therefore, the majority of our software development takes place in a monorepo, and our Python ecosystem is set up such that internal…
joinly
Build personalized, conversational meeting agents in minutes
https://github.com/joinly-ai/joinly
Build personalized, conversational meeting agents in minutes
https://github.com/joinly-ai/joinly
GitHub
GitHub - joinly-ai/joinly: Make your meetings accessible to AI Agents
Make your meetings accessible to AI Agents. Contribute to joinly-ai/joinly development by creating an account on GitHub.
OpenPipe / ART
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, Kimi, and more!
https://github.com/OpenPipe/ART
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, Kimi, and more!
https://github.com/OpenPipe/ART
GitHub
GitHub - OpenPipe/ART: Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on…
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, and more! - OpenPipe/ART
Why Python's deepcopy() is surprisingly slow (and better alternatives)
https://www.reddit.com/r/Python/comments/1mehrc0/why_pythons_deepcopy_is_surprisingly_slow_and/
https://www.reddit.com/r/Python/comments/1mehrc0/why_pythons_deepcopy_is_surprisingly_slow_and/
Reddit
From the Python community on Reddit: Why Python's deepcopy() is surprisingly slow (and better alternatives)
Explore this post and more from the Python community
Llmswap – Python package to reduce LLM API costs by 50-90% with caching
https://pypi.org/project/llmswap
https://pypi.org/project/llmswap
microsoft / MoGe
CVPR'25 Oral MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision
https://github.com/microsoft/MoGe
CVPR'25 Oral MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision
https://github.com/microsoft/MoGe
GitHub
GitHub - microsoft/MoGe: [CVPR'25 Oral] MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal…
[CVPR'25 Oral] MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision - microsoft/MoGe