Building generalist agents that can handle diverse tasks and evolve themselves across different environments is a long-term goal in the AI community. Large language models (LLMs) are considered a promising foundation to build such agents due to their generalized capabilities. Current approaches either have LLM-based agents imitate expert-provided trajectories step-by-step, requiring human supervision, which is hard to scale and limits environmental exploration; or they let agents explore and learn in isolated environments, resulting in specialist agents with limited generalization. In this paper, we take the first step towards building generally-capable LLM-based agents with self-evolution ability. We identify a trinity of ingredients: 1) diverse environments for agent exploration and learning, 2) a trajectory set to equip agents with basic capabilities and prior knowledge, and 3) an effective and scalable evolution method. We propose AgentGym, a new framework featuring a variety of environments and tasks for broad, real-time, uni-format, and concurrent agent exploration. AgentGym also includes a database with expanded instructions, a benchmark suite, and high-quality trajectories across environments. Next, we propose a novel method, AgentEvol, to investigate the potential of agent self-evolution beyond previously seen data across tasks and environments. Experimental results show that the evolved agents can achieve results comparable to SOTA models. We release the AgentGym suite, including the platform, dataset, benchmark, checkpoints, and algorithm implementations. The AgentGym suite is available on https://github.com/WooooDyy/AgentGym.
Learning on the Job: An Experience-Driven Self-Evolving Agent for Long-Horizon Tasks
Central South University（中南大学）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Fudan University（复旦大学）；Shanghai Innovation Institute（上海创新研究院）；Zhejiang University（浙江大学）
ERSkill: Evolving for Skill-Guided Adaptive Memory Retrieval
Shenzhen International Center for Industrial and Applied Mathematics（深圳国际工业与应用数学中心）；Shenzhen Research Institute of Big Data（深圳大数据研究院）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Shenzhen Campus of Sun Yat-sen University（中山大学深圳校区）；Shenzhen Loop Area Institute（深圳河套学院）
PhotoCraft: Agentic Reasoning with Hierarchical Self-Evolving Memory for Deep Image Search
Pattern Recognition Center, WeChat AI, Tencent Inc.（微信AI模式识别中心，腾讯公司）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Zhongguancun Academy（中关村学院）；Nanyang Technological University（南洋理工大学）；Shanghai Jiao Tong University（上海交通大学）
EvoFlow: Evolving Diverse Agentic Workflows On The Fly
The Chinese University of Hong Kong（香港中文大学）；Tongji University（同济大学）；Wuhan University（武汉大学）；Tsinghua University（清华大学）；Nanyang Technological University（南洋理工大学）；Shanghai AI Laboratory（上海人工智能实验室）
AgentStream: How Well Do Self-Evolving LLM Agents Perform Under Streaming Tasks?
University of Chinese Academy of Sciences（中国科学院大学）；Microsoft（微软公司）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Nanjing University（南京大学）
MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents
Nanyang Technological University（南洋理工大学）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of Illinois Chicago（伊利诺伊大学芝加哥分校）；Tsinghua University（清华大学）
Skills-Coach: A Self-Evolving Skill Optimizer via Training-Free GRPO
University of Chinese Academy of Sciences（中国科学院大学）；East China Normal University（华东师范大学）；Zhongguancun Academy（中关村学院）；Southeast University（东南大学）；Hainan University（海南大学）；Tsinghua University（清华大学）
Multi-Agent Collaboration via Evolving Orchestration
Tsinghua University（清华大学）；Shanghai Jiao Tong University（上海交通大学）；Beijing University of Posts and Telecommunications（北京邮电大学）；Siemens（西门子）；Tencent Robotics X（腾讯机器人研究院）
DéjàQ: Open-Ended Evolution of Diverse, Learnable and Verifiable Problems
Willem Röpke AI Lab, Vrije Universiteit Brussel Belgium（维尔姆·罗普克人工智能实验室，布鲁塞尔自由大学）；FLAIR, University of Oxford United Kingdom（FLAIR，牛津大学）；University College London United Kingdom（伦敦大学学院）
Evolving Jailbreaks: Automated Multi-Objective Long-Tail Attacks on Large Language Models
School of Artificial Intelligence（人工智能学院）；Brain-Inspired Technology Co.,Ltd（脑启发技术有限公司）；Department of Computer Science and Engineering（计算机科学与工程系）；Southern University of Science and Technology（南方科技大学）；Nanyang Technological University（南洋理工大学）
Automated Annotation of Evolving Corpora for Augmenting Longitudinal Network Data: A Framework Integrating Large Language Models and Expert Knowledge
Wangxuan Institute of Computer Technology, Peking University（北京大学王选计算机研究所）；School of International Studies and Institute for Carbon Neutrality, Peking University（北京大学国际关系学院与碳中和研究院）；Analytics Lab for Global Risk Politics, Peking University（北京大学全球风险政治分析实验室）
Evolving Medical Imaging Agents via Experience-driven Self-skill Discovery
Southwest Jiaotong University, China（西南交通大学）；The University of Tokyo, Japan（东京大学）；RIKEN, Japan（日本理化学研究所）；Westlake University, China（西湖大学）；Hokkaido University, Japan（北海道大学）
AutoMaAS: Self-Evolving Multi-Agent Architecture Search for Large Language Models
Department of Software \& Microelectronics Peking University Beijing, China；Department of Software \& Microelectronics Peking University Beijing, China hangli\；Department of Software \& Microelectronics Peking University Beijing, China zehua\；Department of Software \& Microelectronics Peking University Beijing, China xiaofan\；Economic Law School China University of Political Science；School of Computer Science Peking University Beijing, China
