×
Open Source

Alibaba Cloud Releases RCA Benchmark, the Industry's First Open Source Root Cause Analysis Benchmark System for Agentic Ops

This article introduces Alibaba Cloud's open-source RCA Benchmark for evaluating AI agents in IT operations.

Apache RocketMQ 5.5.0 Open Source LiteTopic: Dedicated Channel for Millions of AI Sessions

This article introduces Apache RocketMQ 5.5.0's LiteTopic, a new message model designed for millions of lightweight AI agent sessions with event-driven distribution and session persistence.

AgentScope Java 2.0: Building a Distributed, Enterprise-Grade Foundation for AI Agents

This article introduces AgentScope Java 2.0, an open-source framework for building distributed, enterprise-grade AI agents with production-ready features.

SkillClaw × Nacos: Autonomic Evolutionary Closed Loop from an Agent Session to a Governable Skill Registry

This article introduces how SkillClaw and Nacos combine to create a closed loop for AI Agent skills—from automatic generation in real sessions to governed team-wide distribution.

When DuckDB Embraces MySQL in the AI Era

AliSQL enhances MySQL with DuckDB-powered analytical instances for high-performance HTAP capabilities while maintaining full MySQL compatibility.

From an 18-Year-Old Hidden Nginx Vulnerability to the Evolution of Gateway Security Architecture

This article analyzes the root causes of the 18-year-old Nginx vulnerability (CVE-2026-42945) and explores the evolution of gateway security architecture.

Alibaba Cloud Tair KVCache Simulation Analysis: High-Precision Computational and Caching Simulation Design and Implementation

This article introduces Tair-KVCache-HiSim, a high-fidelity CPU-based simulator for optimizing multi-tier KV Cache configurations in LLM inference.

SGLang Hierarchical Sparse Attention

This article introduces hierarchical sparse attention: the full KV Cache is stored on the CPU, while the GPU keeps only a Top-k LRU Buffer.

Hybrid Model Support | SGLang's Support Scheme for Hybrid Architecture Models like Mamba-Transformer

This article introduces a dual memory-pool inference framework enabling efficient hybrid Transformer-Mamba model execution by resolving conflicting caching mechanisms.

Alibaba Cloud Tair KVCache Manager: Architecture Design and Implementation of Enterprise-Level Global KVCache Management Service

This article introduces the architecture and implementation of Tair KVCache Manager, an open-source enterprise-grade global KVCache management service for scalable Agentic AI inference.

Alibaba Cloud Tair KVCache Implementation Based on 3FS Enterprise-Grade Deployment, High-Availability Operations & Performance Optimization

This article introduces engineering optimizations to 3FS—KVCache's foundation layer—across performance, productization, and cloud-native management for scalable AI inference.

The First Java Harness Framework Is Here | AgentScope Brings OpenClaw to Enterprise Distributed Scenarios

AgentScope Java 1.1 launches with workspace-driven persistence, pluggable filesystems, auto-context management, and secure sandbox orchestration for scalable enterprise Agents.

AliSQL DuckDB: Data Compression and Archiving Analysis Practices

This article explains how AliSQL natively supports high-density storage and efficient analysis by deeply integrating DuckDB while maintaining compatibility with the MySQL ecosystem.

When MySQL Meets the Columnar Storage Engine DuckDB in the AI Era

AliSQL integrates DuckDB as a storage engine to add high-performance OLAP capabilities to MySQL while maintaining full compatibility.

Skills Registry Public Beta Launches: Building a Private Skill Management Center for Enterprises

This article introduces Skills Registry, Alibaba Cloud's enterprise-grade private repository for securely managing, versioning, and controlling AI Skills.

FlashQLA: CP-/Bwd-Friendly Fused Linear Attention Kernels for GDN

We officially open-source FlashQLA: a high-performance linear attention kernel library built on TileLang.

Qwen-Scope: Decoding Intelligence, Unleashing Potential

We are excited to introduce Qwen-Scope, an interpretability toolkit trained on the Qwen3 and Qwen3.5 series models.

HiClaw Releases v1.1.0, Delivering a Kubernetes Cluster Deployment Implementation and Support for the Hermes Worker Runtime

This article introduces the new features and bug fixes of HiClaw v1.1.0.

HiClaw: Unified Deployment for OpenClaw and Hermes Workflows

This article introduces a community-driven comparison of OpenClaw and Hermes AI agents, positioning HiClaw as the unified platform to run both without compromise.

Alibaba Open-sources Qwen3.6-35B-A3B; Wan2.7 Tops Design Arena

This week, Alibaba open-sourced Qwen3.6-35B-A3B, a sparse Mixture of Experts (MoE) model that delivers strong performance in agentic coding and complex reasoning following the launch of Qwen3.