Who uses aiologic? #11
Replies: 32 comments 3 replies
|
x42005e1f/culsans (2024-11-02 – present; active) This is my second project that provides an alternative to Janus queues. Initially, Culsans was created because of performance issues in the latter as a proof-of-concept, but over time it grew into an independent project. It relies on The Culsans queues differ from the complex aiologic queues in the following aspects:
The price of these capabilities is that Culsans relies on a thread-level lock, which deviates from the aiologic concept. |
|
MatthewScholefield/meross-auto-lighting (2025-01-20 – present; inactive) A script that controls Meross smart lights. It uses |
|
Vizonex/aiothreading (2025-02-05 – present; active) A fork of omnilib/aiomultiprocess that uses threads instead of processes. Once, I was looking at what solutions had been made before me and came across this project, which coincidentally was published in the same season as mine. It implemented its primitives, but via polling, which was used throughout the entire codebase, and I wanted to comment on a possible improvement. It all started with one issue and ended with the fact that aiothreading now actively relies on aiologic, and I am a collaborator there. It uses various aiologic primitives to assemble asyncio threads and asyncio thread pools. Support for AnyIO is planned. |
|
logicblocks/event.store (2025-02-19 – present; active) An eventing infrastructure with asyncio support that provides various storage adapters. It uses |
|
DmitryBugaychenko/spade-llm (2025-04-01 – present; inactive) A multi-agent system for LLM agents, running on asyncio. It uses |
|
apache/airflow (2025-04-07 – 2025-06-17) A platform for managing workflows with asyncio support. It used The original issue about deadlock with This project is the reason why aiologic has a large number of "dependent" repositories and a large number of downloads in the past. That is why aiologic was in the top 2,000 most-downloaded packages from PyPI. |
|
academy-agents/academy (2025-06-25/2025-07-05 – present; active) A multi-agent system with asyncio support, built on the actor model. It uses |
|
Vizonex/aiointernal (2025-08-06 – present; inactive) A worker thread implementation with an asyncio interface, which delegates the execution of selected methods. It uses |
|
AetherBreaker/LegendLoreBot (2025-08-11 – present; inactive) A Discord bot written using disnake with a modular architecture, running on asyncio. It uses |
|
AetherBreaker/ScheduledOrderMiddleman (2025-10-01 – present; active) A scheduler that executes commands via SFTP (actually, it is something more niche, but I did not go into details), running on asyncio. It uses |
|
fleming79/async-kernel (2025-11-04 – present; active) An implementation of a Python Jupyter kernel with concurrent request handling, running on AnyIO. It uses various aiologic primitives to synchronize asynchronous code with synchronous threads. This project was the reason for the early release of aiologic 0.15.0 (as it was the first project I found that works on something other than asyncio). |
|
microsoft/agent-lightning (2025-12-05 – present; inactive) An optimizing trainer for AI agents. It uses |
|
datarobot-community/talk-to-my-data-agent (2025-12-08 – present; active) A question answering (QA) system over structured data, running on asyncio. It uses |
|
klaw-python/klaw (2025-12-17 – present; inactive) klaw-result is a collection of various types inspired by Rust, with support for AnyIO. It uses |
|
orangeqs/juice (2025-11-21 – present; active) A platform for managing and automating quantum device testing. It uses |
|
AmritaBot/AmritaCore (2026-04-12 – present; active) An agent framework, built on asyncio. It uses |
|
AmritaBot/AmritaBot (2026-04-26 – present; active) A chatbot platform based on AmritaBot/AmritaCore. It uses |
|
AetherBreaker/sft_ext (2026-05-21 – present; active) A shared library containing all the primitives whose usage was described in relation to the same author's projects (see above). |
|
AmritaBot/AmritaSense (2026-05-17 – present; active) An interpretive workflow engine, running on asyncio. It encapsulates some classes previously found in AmritaCore, such as |
|
AmritaBot/plugin-memory (2026-05-02 – present; active) … |
|
eeeob/aldheeb-pytools (2026-06-02 – 2026-07-30) … |
|
WePositive-energy/wepositive-di (2026-06-03 – present; active) … |
|
Vizonex/anyio-executor (2026-06-10 – present; active) … |
|
LiteSuggarDEV/nonebot_plugin_amrita (2026-06-21 – present; active) … |
|
AetherBreaker/shared-log-processor (2026-07-01 – present; unavailable) … |
|
iyad-f/sonyflake (2026-07-07 – present; active) … |
|
Attacking-Lab/ctf-attackapi (2026-02-07 – present; active) … |
|
AmritaBot/mcp-ssh-reloaded (2026-07-13 – present; active) … |
|
redsun-acquisition/redsun (2026-07-26 – present; active) … |
|
AmritaBot/Faskill (2026-08-03 – present; active) … |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
There are three kinds of lies: lies, damned lies, and statistics. The same applies in our case: stars indicate how popular the idea might be (how successful the project's promotion is; even star-history.com recommends to inflate the star count!), downloads reflect the load on CI/CD (in particular, from major dependents) and forks generally do not tell us anything. We cannot even rely on feedback, since for high-quality tools… what is there to talk about anyway (if you put the tool under the hood and it does not bother you in any way, you will easily forget it even exists)? There are projects with hundreds of stars that are sometimes discussed in chat rooms but actually have zero usage, which just confirms how absurd these metrics are.
What if we want to measure the usefulness or usability of the project? We could analyze the dependents, but there is a problem: GitHub provides information on all dependents, including indirect ones (which are the vast majority; if A uses B, and B uses C, there is transitive "A uses C", but A itself knows nothing about C), and has false negatives and false positives. That is why this discussion exists.
Here, I will talk about direct dependents that I found on the Internet and how they use aiologic. I invite everyone who wants to join, especially those who use the project privately or on platforms where it is difficult to find any information (such as GitLab).
Last checked: 2026-08-07.
All reactions