# tsinghua-fib-lab/AgentSociety

AgentSociety 2 is a modern, LLM-native agent simulation platform designed for social science research and experimental design. It provides a flexible framework for creating and managing intelligent agents in simulated environments.

Repository: https://github.com/tsinghua-fib-lab/AgentSociety
Canonical: https://ross.abutalabs.com/products/agentsociety
Homepage: https://agentsociety2.fiblab.net
Language: Python
License: Apache-2.0
License Family: permissive
Topics: agentsociety2, ai-social-scientist, agent, multi-agents, simulation, agent-based-modeling, ai-assistants, ai-for-science, harness-engineering, social-simulation
Last push: 2026-08-18T16:56:24+00:00

## Health v2 (maintenance only)
Score: 85/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 95, longevity 40
- inputs: {"age_days": 573, "days_push": 15, "days_rel": 37, "gap_med": 7.0, "n_releases_24m": 23}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1236, forks 207 (observed 2026-08-28T04:04:05.029915+00:00)

## What it is
AgentSociety is an LLM-native agent simulation framework for building large-scale multi-agent simulations of social and urban environments. It supports social science research with pluggable environments, multiple agent reasoning patterns, and experiment workflows.

## Use cases
- simulate a society of LLM agents in a city environment
- run agent-based modeling experiments for social science research
- build multi-agent simulations with LLM-driven reasoning
- test social science hypotheses with simulated populations
- compare agent reasoning patterns like ReAct and plan-execute
- model urban mobility and crowd behavior with intelligent agents

## When to choose
- you need LLM-driven multi-agent social or urban simulations
- you are a researcher doing computational social science experiments
- you want a Python framework with pluggable environments and reasoning patterns

## When to avoid
- you need lightweight, non-LLM agent-based modeling at massive scale
- you want a general-purpose game or robotics simulator
- you need production multi-agent orchestration for business workflows

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, simulation, llm-inference, machine-learning
- domain: artificial-intelligence, simulation, social-media
- platform: python, cross-platform
- tags: llm-agents, social-simulation, agent-based-modeling, social-science, urban-simulation, multi-agent-systems, ai-agents, research

## Member repositories
- tsinghua-fib-lab/AgentSociety (main) score 85

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:05.029915+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T08:22:01.553118+00:00, confidence not recorded.
  - readme: https://github.com/tsinghua-fib-lab/AgentSociety (fetched 2026-08-28T04:04:05.029915+00:00, sha 05da7f1eee38)
  - homepage: https://agentsociety2.fiblab.net (fetched 2026-08-29T12:21:38.388459+00:00, sha 6327e34db2b1)
  - registry_pypi: https://pypi.org/pypi/agentsociety/json (fetched 2026-08-29T12:21:38.391527+00:00, sha 5c5605856d07)
- Data as of 2026-08-30T08:39:29.467469+00:00.