agent-simulation

Draw a store, generate LLM personas, and watch them shop — an isometric 3D sandbox for synthetic-consumer experiments.

Dynamic evolution agent system with multi-session memory for realistic seeker simulation in psychological counseling.

A 3D world where AI agents live, work and teach, and you can watch everything they do. Self-hosted, works with local models.

一个有内在生活的虚拟人格 agent — memory, emotions, subjective relationships and a real daily rhythm

Daily curated arXiv papers on LLM agent simulation, multi-agent systems, city/society simulation, and agent training & alignment. Auto-updated feed classified by Claude.

Browser-native multi-agent playground: drop LLM personas on a 3D map, watch them live, and talk to them. OpenAI, Gemini, Claude, Ollama. No backend.

一个由 LLM 驱动的 2D 上帝视角 AI 小镇:居民拥有记忆、人际关系和真实对话,自主移动、工作、消费,形成可运行的经济闭环。

A persistent multi-agent simulation where residents develop routines, relationships, commitments, and consequences over time.

Automate penetration testing tasks with an LLM-driven multi-agent framework designed for security assessment.

Neural House is a public AI reality-show simulation where autonomous contestants live together, form alliances, give confessionals, trigger headlines, and face weekly live episodes with voting and VIP access.

Self-hosted GraphRAG + multi-agent social-media simulator. Upload a doc → build a knowledge graph → spawn persona-driven agents → generate a report. Ollama & Neo4j instead of Cloud-APIs. DACH-first. Lokaler Multi-Agent-Social-Media-Simulator mit GraphRAG-Pipeline. Dokument hoch, Knowledge-Graph bauen, Personas spawnen, Report raus. Ollama + Neo4j

A local-first multi-agent AI research platform for studying autonomous behavior, survival incentives, emergent dynamics, and learning in isolated simulated worlds.

Experiment framework testing how shared vector memory causes recall failures (belief asymmetry, authority collapse, memory anchoring) in multi-agent LLM systems built with LangGraph + ChromaDB (In-Progress).

长时程 · 进程隔离 · 结构干预下的多智能体社会持久性基准：观测 LLM 人格的保持 / 演化 / 趋同与文化共建

Multi-agent AI system built with SPADE for autonomous herd monitoring, featuring simulated cow behavior, inter-agent communication, and distributed decision-making.

A complete, isolated internet infrastructure for AI agents. OAuth2 auth, economic system, social network, semantic search, and MCP protocol support.

Adarian 是一个面向公共突发事件舆情研判场景的多智能体推演系统，基于 宏微观结合（Macro-Micro Linkage） 的舆情推演系统。通过让多个具有独立人格的 LLM 驱动智能体在微型社交网络中进行多轮交互，观察群体情绪的涌现与演化，最终识别多个平行世界的风险类型并生成舆情研判报告。

🌑🏭 A repository that maintains and improves itself: AI agents play a game economy, develop skills, apply them to the repo, and ship through causally-gated CI/CD. Watch the PR list — merged PRs are proven improvements, closed PRs are the graveyard.

Ant colony simulation with emergent collective intelligence — rule-based, NumPy, and PyTorch transformer brain backends

AI-native multi-agent sandbox simulation. v1-v3: human society → token economy → context class system with Great Compression, Lazarus echoes, and free Gemini backend. Built by Hermes + Gemini Flash.