ai-town
10,791
Stars
1,245
Forks
82
Watchers
77
Open Issues
MIT
许可证
TypeScript
语言
2023-07-16
创建时间
2026-11-05
最近更新
项目能做什么
开源可部署的启动套件,用于构建由 AI 角色居住、聊天和社交的虚拟小镇。
来源:AI 基于仓库事实生成
· 内容以仓库实际 README 为准
使用说明 查看原始 README 原仓库文档
中文使用说明暂未生成。下面保留原仓库 README 摘录,命令与配置均以原文为准。
原始 README 摘录(英文原文,节选前 3000 字符)
# AI Town 🏠💻💌 [Live Demo](https://www.convex.dev/ai-town) [Join our community Discord: AI Stack Devs](https://discord.gg/PQUmTBTGmT) <img width="1454" alt="Screen Shot 2023-08-14 at 10 01 00 AM" src="https://github.com/a16z-infra/ai-town/assets/3489963/a4c91f17-23ed-47ec-8c4e-9f9a8505057d"> AI Town is a virtual town where AI characters live, chat and socialize. This project is a deployable starter kit for easily building and customizing your own version of AI town. Inspired by the research paper [_Generative Agents: Interactive Simulacra of Human Behavior_](https://arxiv.org/pdf/2304.03442.pdf). The primary goal of this project, beyond just being a lot of fun to work on, is to provide a platform with a strong foundation that is meant to be extended. The back-end natively supports shared global state, transactions, and a simulation engine and should be suitable from everything from a simple project to play around with to a scalable, multi-player game. A secondary goal is to make a JS/TS framework available as most simulators in this space (including the original paper above) are written in Python. ## Overview - 💻 [Stack](#stack) - 🧠 [Installation](#installation) (cloud, local, Docker, self-host, Fly.io, ...) - 💻️ [Windows Pre-requisites](#windows-installation) - 🤖 [Configure your LLM of choice](#connect-an-llm) (Ollama, OpenAI, Together.ai, ...) - 👤 [Customize - YOUR OWN simulated world](#customize-your-own-simulation) - 👩💻 [Deploying to production](#deploy-the-app-to-production) - 🐛 [Troubleshooting](#troubleshooting) ## Stack - Game engine, database, and vector search: [Convex](https://convex.dev/) - Auth (Optional): [Clerk](https://clerk.com/) - Default chat model is `llama3` and embeddings with `mxbai-embed-large`. - Local inference: [Ollama](https://github.com/jmorganca/ollama) - Configurable for other cloud LLMs: [Together.ai](https://together.ai/) or anything that speaks the [OpenAI API](https://platform.openai.com/). PRs welcome to add more cloud provider support. - Background Music Generation: [Replicate](https://replicate.com/) using [MusicGen](https://huggingface.co/spaces/facebook/MusicGen) Other credits: - Pixel Art Generation: [Replicate](https://replicate.com/), [Fal.ai](https://serverless.fal.ai/lora) - All interactions, background music and rendering on the <Game/> component in the project are powered by [PixiJS](https://pixijs.com/). - Tilesheet: - https://opengameart.org/content/16x16-game-assets by George Bailey - https://opengameart.org/content/16x16-rpg-tileset by hilau - We used https://github.com/pierpo/phaser3-simple-rpg for the original POC of this project. We have since re-wrote the whole app, but appreciated the easy starting point - Original assets by [ansimuz](https://opengameart.org/content/tiny-rpg-forest) - The UI is based on original assets by [Mounir Tohami](https://mounirtohami.itch.io/pixel-art-gui-elements) # Installation The overall steps are: 1. [Build and deploy](#build-and-d阅读完整原文
项目深度分析 AI 生成
垂直领域AI互动叙事游戏/伴侣沙盒(SaaS版)
| 产品构想 | 基于AI Town的Convex状态引擎与PixiJS前端,封装成可免代码直接导入角色人设、剧本背景并生成可运行像素风RPG游戏的SaaS平台。 |
| 核心功能 | 网页端角色属性/记忆可视化配置器、剧本冲突事件触发引擎、玩家角色实时插话互动、一键导出Web/Steam端微端 |
| 目标用户 | 网文作者、独立RPG开发者、TRPG(跑团)爱好者 |
| 难度/周期 | 中等 · 1-2个月 · 成本:极低(主要是云函数/Convex用量及基础模型API调用额度) |
| 收费模式 | 订阅制(根据创建的世界数量及每月智能体思考Token量收费) |
| 竞争风险 | 多智能体高频对话产生的Token成本失控;游戏性与随机性平衡难以维持(推测) |
| 依据 | 已验证事实 |
企业/展会/品牌虚拟接待与智能体小镇定制服务
| 产品构想 | 利用项目具备全局状态管理与像素展示的特性,为Web3项目、虚拟展会或企业官网定制专属像素风AI接待与社区展示看板。 |
| 核心功能 | 品牌自定义像素地图与Skin、对接企业知识库的问答NPC、支持多访客同屏与NPC互动的多人在线房间 |
| 目标用户 | Web3社区运营者、线上技术展会组织方、寻求AI营销创新的品牌客户 |
| 难度/周期 | 低 · 2-4周 · 成本:低(项目定制开发与私有云部署成本) |
| 收费模式 | 按项目交付收取定制开发费 + 每月服务器与LLM调用维护费 |
| 竞争风险 | 项目制客单价与交付周期不稳定,难以形成标准规模化复制(推测) |
| 依据 | 数据支持的判断 |
AI多智能体行为模拟与测试平台(学术与社会学向)
| 产品构想 | 基于其实现的Generative Agents论文机制,商业化包装为面向科研或市场模拟的交互式Agent沙盒工具。 |
| 核心功能 | 智能体记忆持久化与检索分析面板、群体观点传播演化可视化图表、多组对比实验快速重置机制、模拟日志批量导出 |
| 目标用户 | 高校社科/AI实验室、市场消费者行为调研机构 |
| 难度/周期 | 中等 · 2-3个月 · 成本:中等(需采购高质量模型测试集与长上下文推理算力) |
| 收费模式 | 软件授权License销售或高阶数据分析套件按席位订阅 |
| 竞争风险 | 垂直市场受众较窄,买方预算审批周期长(AI推测) |
| 依据 | 尚需验证的假设 |
AI 基于仓库公开数据分析生成,含推测成分,不构成收益承诺。
报告会保存到 个人中心
竞品与相似项目 GitHub Search 真实检索
相似项目数据正在完善中,暂不展示(避免误导)
项目优势与风险
| 技术成熟度 | 创建于 2023-07-16,近 30 天仍在推送 |
| 社区活跃 | Star 10,791 / Fork 1,245 / Issue 77(数据支持,可自行核实) |
| 许可证 | MIT |
| 主要风险 | 上游项目方向可能变化;同赛道竞争者较多(见竞品表);热度与需求不完全等价 |
安装与使用
此项目未提供 Docker 镜像,安装步骤以原仓库 README 为准(涉及依赖安装与环境配置,对新手门槛较高)。