Open source AI
11 notes, newest first.
UniRL: A Framework for Unified Multimodal Model Reinforcement Learning
Tencent Hunyuan released UniRL, an open-source framework that applies one shared reinforcement-learning training loop across different kinds of generative models (diffusion/image models, text language models…
qwen-code-docs: documentation translation tool for the Qwen Code project
Alibaba's Qwen team released a small open-source tool that automatically translates the Qwen Code project's documentation into eight languages and builds a multilingual documentation website, using Qwen's own AI models…
qwen-code v0.21.12 release (and related preview/nightly builds)
Alibaba's Qwen team shipped a stable release (v0.21.12) of qwen-code, its coding-agent CLI/desktop tool, along with preview and nightly builds in the same window, adding features like drag-and-drop workspace file…
HiLS-Attention: Hierarchical Sparse Attention Done Right
Tencent's Hunyuan team released official code for HiLS-Attention, a chunk-based sparse attention method meant to make very long-context language models faster to run while keeping accuracy, learning which chunks of…
Hunyuan3D-WorldClaw: Agentic 3D Open-world Generation at Scale
Tencent's Hunyuan team released "WorldClaw," a system that uses AI agents combined with procedural generation to build large open-world 3D environments.
CheMLFlow: An Open-Source Platform for Cheminformatics and Materials Informatics Applications
CheMLFlow is an open-source platform designed to facilitate the creation and execution of end-to-end, high-throughput workflows for scientific and technological applications. It aims to streamline the process of…
A First Look at Coding Agents' Compliance with AI Contribution Rules in Open-Source Communities
Researchers studied whether AI agents comply with contribution rules in open-source communities. They found that the AI agents do not proactively retrieve the rules but improve with reminders, quotes, or feedback…
GigaPath-Flash and GigaTIME-Flash: Efficient Pathology Foundation Models for Whole-Slide and Tumor Microenvironment Analysis
Researchers have developed two new foundation models called GigaPath-Flash and GigaTIME-Flash for analyzing whole-slide pathology and predicting the tumor immune microenvironment from routine H&E images. These models…
MedFailBench: A Clinician-Built Open-Source Benchmark for Medical AI Safety Boundary Inspection
MedFailBench is a new open-source benchmark for evaluating the safety of medical AI systems, created by clinicians. It focuses on identifying where a system has failed in terms of safety boundaries rather than just…
Early Adoption of Agentic Coding Tools by GitHub Projects
This research investigates the adoption and management of agentic coding tools in software development on GitHub. The study analyzes 25,264 agentic pull requests from 2,361 popular repositories to understand (1) how…
Danus: Orchestrating Mathematical Reasoning Agents with Fact-Graph Memory
The paper introduces Danus, a system designed to effectively coordinate and scale mathematical reasoning agents by using a shared fact graph as a global memory-management mechanism. The system consists of a main agent…
