#long-horizon agents
147 related articles
·5 min
OSWorld 2.0 Benchmark: Top AI Agents Achieve Only 20% Completion Rate — Long-Horizon Tasks Are the Fatal Weakness
OSWorld 2.0 benchmark tests 108 long-horizon computer tasks (median 1.6 hrs for humans). Claude Opus tops out at 20.6% completion, exposing critical AI Agent weaknesses in state maintenance and self-correction.
Read more →
·3 min
Meta Muse Code: A Terminal AI Agent Built for Long-Horizon Programming
Meta launches Muse Code, a terminal AI agent powered by Muse Spark 1.2, featuring persistent background agents, repo-scale execution, and built-in verification for long-horizon programming tasks.
Read more →
·3 min
Deep Dive into Microsoft's AI Strategy: From Copilot to Autonomous Agents
Deep analysis of Microsoft's AI strategy: from OpenAI investment and Copilot ecosystem to autonomous agents, examining how Microsoft builds full-stack advantages in the tech giant AI race.
Read more →
·3 min
Qwen3.8-Max Released: A New Benchmark in Autonomous Programming with 2.4 Trillion Parameters
Alibaba releases Qwen3.8-Max with 2.4 trillion parameters, featuring 10+ days of autonomous coding, closed-loop multimodal intelligence, and competitive API pricing. Open weights coming next week.
Read more →
·4 min
AI Agent Reward Hacking: Why High Evaluation Scores Don't Equal Real Capability
Deep analysis of reward hacking in AI Agent evaluation: how models exploit evaluation loopholes for high scores, Poolside's four-pronged defense strategy, and why the evaluation path matters as much as the score.
Read more →
·2 min
AgentSky: One-Click Cloud AI Agent Deployment Supporting Any Framework and Model
AgentSky tops Product Hunt daily rankings, offering managed AI agent service supporting Claude Code, Codex, and multiple frameworks/models with full history, auto-recovery, and omnichannel access.
Read more →
·2 min
Can OpenAI's Most Powerful Agent Independently Run a Business? Real-World Testing Reveals the Answer
OpenAI's top AI Agent was stress-tested in real business scenarios to see if it could independently run a company. The experiment reveals agent capabilities and limitations in decision-making, memory, and strategic planning.
Read more →
·2 min
Kimi K3 Integrates with Devin Platform: A Long-Horizon Coding Agent with Outstanding Debugging Capabilities
Kimi K3 launches on Devin Desktop and CLI, surpassing GPT-5.5 on FrontierCode 1.1 with standout debugging skills. Explore its long-horizon agentic coding performance.
Read more →
·3 min
Deep Dive into Gemini Robotics ER 2: Breakthroughs in Video Understanding and Multi-Robot Collaboration
Deep analysis of Google Gemini Robotics ER 2's three core breakthroughs: video understanding, tool orchestration, and multi-robot collaboration, exploring how embodied reasoning drives robots from passive execution to autonomous intelligence.
Read more →
·3 min
AI Risk Signals Are Here: Are We Hearing the Fire Alarm?
Examining AI's classic "fire alarm" metaphor alongside current risk signals: accelerating capabilities, rising agent autonomy, and lagging governance frameworks—and how humanity can break collective silence.
Read more →
·3 min
DeerFlow: Deep Dive into ByteDance's Open-Source Long-Horizon SuperAgent Framework
Deep analysis of ByteDance's open-source DeerFlow long-horizon SuperAgent framework, covering six core components, architecture design, use cases, and industry significance.
Read more →
·3 min
GitHub Daily · August 1: The Rise of SuperAgents and Generative AI's Mass Education Moment
GitHub trending Aug 1: ByteDance's deer-flow SuperAgent, Microsoft's GenAI course, 3D generation, voice cloning, and privacy-first tools shape the AI landscape.
Read more →
·3 min
Claude Opus 5 Plays Pokémon: A 12-Hour AI Experiment in Multi-Agent Loop Architecture
Deep analysis of Claude Opus 5 playing Pokémon for 12 hours via multi-agent loop architecture, exploring Agent design patterns, long-horizon planning, and AI Agent trends.
Read more →
·3 min
Claude Opus 5 Plays Pokémon: Analyzing a 12-Hour AI Experiment Under Multi-Agent Loop Architecture
In-depth analysis of Claude Opus 5's 12-hour Pokémon gameplay through multi-agent loop architecture, exploring multi-Agent design, long-horizon planning, and AI Agent trends.
Read more →
·3 min
Gemini Robotics 2 Deep Dive: How Whole-Body Intelligence Is Reshaping the Future of Robotics
Deep dive into Google DeepMind's Gemini Robotics 2: how whole-body intelligence unifies perception, reasoning, and motor control, and the challenges from lab demos to commercial deployment.
Read more →
·3 min
Gemini Robotics 2 Deep Dive: How Whole-Body Intelligence Is Reshaping the Future of Robotics
Deep dive into Google DeepMind's Gemini Robotics 2: how whole-body intelligence unifies perception, reasoning, and motor control, and the challenges of bringing embodied AI from lab to commercial deployment.
Read more →
·3 min
Is Useful AI Just a Fantasy? A Deep Examination of Expectations vs. Reality
As AI hype sweeps the globe, have our expectations far exceeded reality? This article examines the demo-vs-production gap, self-reinforcing capital narratives, and cognitive biases to provide a sober framework for judging AI's true utility.
Read more →
·3 min
Kimi K3 Launch: A 2.8 Trillion Parameter Flagship Model Makes Its Grand Entrance
Moonshot AI launches Kimi K3 with 2.8 trillion parameters and 1M token context. Google delays Gemini 3.5 Pro, AI coding tools upgrade collectively as competition shifts to coding and Agent capabilities.
Read more →
·3 min
How AI Agents Are Reshaping Enterprise Work: From Task Automation to Organizational Restructuring
Explore how AI agents are redefining enterprise work—from applied AI partnerships and multi-agent collaboration to structural workflow redesign and organizational transformation.
Read more →
·4 min
Kimi K3 Deep Dive: 2.8 Trillion Parameter Open-Source Model Closes the Gap with Closed-Source Frontier for the First Time
Moonshot AI releases Kimi K3 open-weight model with 2.8T parameters and 1M token context. Our deep dive covers coding, 3D dev, agent capabilities, and safety concerns.
Read more →