Q Labs introduced Dust, a zeroth-order optimization method that estimates update directions from random, token-level activation perturbations instead of standard backpropagation. The company reported training Transformers on FineWeb up to 243M parameters and 20M tokens, while a SOTAAZ Lab reproduction found basic reproducibility but no clear advantage over backpropagation. Dust also remained far more computationally expensive than backpropagation.
What problem it solves
AI agents may exceed their role, take unauthorized actions, or fail to follow organizational goals and policies. Teams need a way to test whether an agent behaves as intended across realistic and adversarial requests.
Key features
Business-aware simulations — Describe an agent, build a simulation environment for it, and exercise the agent with realistic requests that may exc…
What problem it solves
The project brings market monitoring, watchlists, portfolio records, research, news, backtesting, and report management into one workbench across China A-shares, US stocks, and HK stocks. It also keeps research records and imported materials on the user's computer while allowing research to be conducted with local AI subscriptions or compatible model APIs.
Key features
Unifi…
What problem it solves
Developers, researchers, teams, and students need practical examples to understand and evaluate AI agent architectures. The collection organizes production examples, tutorials, and working code by framework and industry to support implementation and learning.
Key features
Runnable agent examples — Each project under agents/ is presented as a self-contained implementation wit…
What problem it solves
Organizations need to store, ingest, and query growing volumes of time series data while supporting the protocols and query languages used by existing monitoring systems. They also need deployment options and operational features that support long-term storage, backups, and centralized querying across data sources.
Key features
Long-term time series storage — Stores time ser…
What problem it solves
Container hosts need a runtime that can manage the complete container lifecycle, including image transfer and storage, container execution and supervision, and low-level storage and network attachments. The runtime must also support integration into larger systems and operation across Linux and Windows.
Key features
Complete container lifecycle management — The daemon manage…
What problem it solves
Developers need a code editor that supports both individual work and multiplayer workflows across major desktop operating systems. Contributors also need clear platform-specific build instructions and a process for adding and maintaining open-source components.
Key features
Multiplayer editing — Zed is positioned as a multiplayer code editor, allowing code editing workflows …