SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2606.07513 · 2026

Agentopia: Long-Term Life Simulation and Learning in Agent Societies

Qi Deng, Yunzhe Tao, Heng Wang, Xintao Wang, Jiawei Wang, Jen-Tse Huang, Minghao Zhu, Sirui Zheng, Xiaojian Wu, Hongqiu Wu, Qianyu He, Weiyuan Li, and 1 more

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 8 functions out of this paper's own repositories and ran 8 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
Neph0s/Agentopia canonical 8 of 8
FunctionStatusWhere it lives
build_god_generate_encounter_events_prompt Ran Neph0s/Agentopia/src/agents/prompts.py
pointer only (licence: NONE) · get_code("556bc8ef821c23a9")
build_god_generate_public_events_prompt Ran Neph0s/Agentopia/src/agents/prompts.py
pointer only (licence: NONE) · get_code("6f10c72b5f05a3ed")
dedupe_tool_calls Ran Neph0s/Agentopia/src/agents/functions.py
pointer only (licence: NONE) · get_code("28cc89d7e126ac5f")
load_config Ran Neph0s/Agentopia/src/config.py
pointer only (licence: NONE) · get_code("fd1f2fb99c1682d5")
split_response_by_visible_blocks Ran Neph0s/Agentopia/src/world/god.py
pointer only (licence: NONE) · get_code("9057eb213b911ec2")
strip_roleplay_instructions Ran Neph0s/Agentopia/src/agents/response_validator.py
pointer only (licence: NONE) · get_code("04dbfdb281faa49c")
validate_activity_tags Ran Neph0s/Agentopia/src/agents/response_validator.py
pointer only (licence: NONE) · get_code("c9c5c57705882aea")
validate_solo_activity_format Ran Neph0s/Agentopia/src/agents/response_validator.py
pointer only (licence: NONE) · get_code("6543f413c07f0517")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Humans learn from social life. Simulating this process with LLM-powered agents represents a promising research direction, raising a natural question: whether LLMs can learn from such simulated social experience to better understand and replicate human behavior. However, prior agent society simulations typically operate at the scale of days, limiting the depth of social interactions and long-term growth. In this paper, we study long-term life simulation and LLM learning in agent societies, with two goals: (1) investigating social behaviors that emerge from life-long simulation, and (2) developing anthropomorphic capabilities in LLMs, particularly intelligence in social life, through years of simulated social experience. Specifically, we present Agentopia, a comprehensive framework for long-term life simulation in multi-agent societies, where 100 agents autonomously pursue personal growth, develop social relationships, and fulfill their needs and goals over 10 simulated years. We define life reward to mirror human well-being, and leverage this reward to train LLMs via rejection sampling. Extensive experiments show that agents exhibit rich emergent social behaviors. Furthermore, life reward training effectively enhances the underlying LLM, which leads to improved agent well-being in simulation, and generalizes to downstream role-playing benchmarks with +15.6% improvement. Our code is available at https://github.com/Neph0s/Agentopia. Figure 1: Illustration of emergent behaviors observed in Agentopia simulations. Each scene depicts a real behavioral pattern documented in our case studies (Tables 22232425262728293031323334). Without explicit scripting, agents autonomously develop diverse behavioral patterns reflecting agents' intelligence in social life.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2606.07513")
get_code_for_paper("2606.07513")
have("2606.07513")

Connect an agent — have() is free.