[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-71224":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":23,"hasPages":23,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":30,"readmeContent":31,"aiSummary":32,"trendingCount":16,"starSnapshotCount":16,"syncStatus":33,"lastSyncTime":34,"discoverSource":35},71224,"Voyager","MineDojo\u002FVoyager","MineDojo","An Open-Ended Embodied Agent with Large Language Models","https:\u002F\u002Fvoyager.minedojo.org\u002F",null,"JavaScript",6968,676,71,1,0,12,26,80,36,105.49,"MIT License",false,"main",[26,27,28,29],"embodied-learning","large-language-models","minecraft","open-ended-learning","2026-06-12 04:00:59","# Voyager: An Open-Ended Embodied Agent with Large Language Models\n\u003Cdiv align=\"center\">\n\n[[Website]](https:\u002F\u002Fvoyager.minedojo.org\u002F)\n[[Arxiv]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.16291)\n[[PDF]](https:\u002F\u002Fvoyager.minedojo.org\u002Fassets\u002Fdocuments\u002Fvoyager.pdf)\n[[Tweet]](https:\u002F\u002Ftwitter.com\u002FDrJimFan\u002Fstatus\u002F1662115266933972993?s=20)\n\n[![Python Version](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPython-3.9-blue.svg)](https:\u002F\u002Fgithub.com\u002FMineDojo\u002FVoyager)\n[![GitHub license](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002FMineDojo\u002FVoyager)](https:\u002F\u002Fgithub.com\u002FMineDojo\u002FVoyager\u002Fblob\u002Fmain\u002FLICENSE)\n______________________________________________________________________\n\n\nhttps:\u002F\u002Fgithub.com\u002FMineDojo\u002FVoyager\u002Fassets\u002F25460983\u002Fce29f45b-43a5-4399-8fd8-5dd105fd64f2\n\n![](images\u002Fpull.png)\n\n\n\u003C\u002Fdiv>\n\nWe introduce Voyager, the first LLM-powered embodied lifelong learning agent\nin Minecraft that continuously explores the world, acquires diverse skills, and\nmakes novel discoveries without human intervention. Voyager consists of three\nkey components: 1) an automatic curriculum that maximizes exploration, 2) an\never-growing skill library of executable code for storing and retrieving complex\nbehaviors, and 3) a new iterative prompting mechanism that incorporates environment\nfeedback, execution errors, and self-verification for program improvement.\nVoyager interacts with GPT-4 via blackbox queries, which bypasses the need for\nmodel parameter fine-tuning. The skills developed by Voyager are temporally\nextended, interpretable, and compositional, which compounds the agent’s abilities\nrapidly and alleviates catastrophic forgetting. Empirically, Voyager shows\nstrong in-context lifelong learning capability and exhibits exceptional proficiency\nin playing Minecraft. It obtains 3.3× more unique items, travels 2.3× longer\ndistances, and unlocks key tech tree milestones up to 15.3× faster than prior SOTA.\nVoyager is able to utilize the learned skill library in a new Minecraft world to\nsolve novel tasks from scratch, while other techniques struggle to generalize.\n\nIn this repo, we provide Voyager code. This codebase is under [MIT License](LICENSE).\n\n# Installation\nVoyager requires Python ≥ 3.9 and Node.js ≥ 16.13.0. We have tested on Ubuntu 20.04, Windows 11, and macOS. You need to follow the instructions below to install Voyager.\n\n## Python Install\n```\ngit clone https:\u002F\u002Fgithub.com\u002FMineDojo\u002FVoyager\ncd Voyager\npip install -e .\n```\n\n## Node.js Install\nIn addition to the Python dependencies, you need to install the following Node.js packages:\n```\ncd voyager\u002Fenv\u002Fmineflayer\nnpm install -g npx\nnpm install\ncd mineflayer-collectblock\nnpx tsc\ncd ..\nnpm install\n```\n\n## Minecraft Instance Install\n\nVoyager depends on Minecraft game. You need to install Minecraft game and set up a Minecraft instance.\n\nFollow the instructions in [Minecraft Login Tutorial](installation\u002Fminecraft_instance_install.md) to set up your Minecraft Instance.\n\n## Fabric Mods Install\n\nYou need to install fabric mods to support all the features in Voyager. Remember to use the correct Fabric version of all the mods. \n\nFollow the instructions in [Fabric Mods Install](installation\u002Ffabric_mods_install.md) to install the mods.\n\n# Getting Started\nVoyager uses OpenAI's GPT-4 as the language model. You need to have an OpenAI API key to use Voyager. You can get one from [here](https:\u002F\u002Fplatform.openai.com\u002Faccount\u002Fapi-keys).\n\nAfter the installation process, you can run Voyager by:\n```python\nfrom voyager import Voyager\n\n# You can also use mc_port instead of azure_login, but azure_login is highly recommended\nazure_login = {\n    \"client_id\": \"YOUR_CLIENT_ID\",\n    \"redirect_url\": \"https:\u002F\u002F127.0.0.1\u002Fauth-response\",\n    \"secret_value\": \"[OPTIONAL] YOUR_SECRET_VALUE\",\n    \"version\": \"fabric-loader-0.14.18-1.19\", # the version Voyager is tested on\n}\nopenai_api_key = \"YOUR_API_KEY\"\n\nvoyager = Voyager(\n    azure_login=azure_login,\n    openai_api_key=openai_api_key,\n)\n\n# start lifelong learning\nvoyager.learn()\n```\n\n* If you are running with `Azure Login` for the first time, it will ask you to follow the command line instruction to generate a config file.\n* For `Azure Login`, you also need to select the world and open the world to LAN by yourself. After you run `voyager.learn()` the game will pop up soon, you need to:\n  1. Select `Singleplayer` and press `Create New World`.\n  2. Set Game Mode to `Creative` and Difficulty to `Peaceful`.\n  3. After the world is created, press `Esc` key and press `Open to LAN`.\n  4. Select `Allow cheats: ON` and press `Start LAN World`. You will see the bot join the world soon. \n\n# Resume from a checkpoint during learning\n\nIf you stop the learning process and want to resume from a checkpoint later, you can instantiate Voyager by:\n```python\nfrom voyager import Voyager\n\nvoyager = Voyager(\n    azure_login=azure_login,\n    openai_api_key=openai_api_key,\n    ckpt_dir=\"YOUR_CKPT_DIR\",\n    resume=True,\n)\n```\n\n# Run Voyager for a specific task with a learned skill library\n\nIf you want to run Voyager for a specific task with a learned skill library, you should first pass the skill library directory to Voyager:\n```python\nfrom voyager import Voyager\n\n# First instantiate Voyager with skill_library_dir.\nvoyager = Voyager(\n    azure_login=azure_login,\n    openai_api_key=openai_api_key,\n    skill_library_dir=\".\u002Fskill_library\u002Ftrial1\", # Load a learned skill library.\n    ckpt_dir=\"YOUR_CKPT_DIR\", # Feel free to use a new dir. Do not use the same dir as skill library because new events will still be recorded to ckpt_dir. \n    resume=False, # Do not resume from a skill library because this is not learning.\n)\n```\nThen, you can run task decomposition. Notice: Occasionally, the task decomposition may not be logical. If you notice the printed sub-goals are flawed, you can rerun the decomposition.\n```python\n# Run task decomposition\ntask = \"YOUR TASK\" # e.g. \"Craft a diamond pickaxe\"\nsub_goals = voyager.decompose_task(task=task)\n```\nFinally, you can run the sub-goals with the learned skill library:\n```python\nvoyager.inference(sub_goals=sub_goals)\n```\n\nFor all valid skill libraries, see [Learned Skill Libraries](skill_library\u002FREADME.md).\n\n# FAQ\nIf you have any questions, please check our [FAQ](FAQ.md) first before opening an issue.\n\n# Paper and Citation\n\nIf you find our work useful, please consider citing us! \n\n```bibtex\n@article{wang2023voyager,\n  title   = {Voyager: An Open-Ended Embodied Agent with Large Language Models},\n  author  = {Guanzhi Wang and Yuqi Xie and Yunfan Jiang and Ajay Mandlekar and Chaowei Xiao and Yuke Zhu and Linxi Fan and Anima Anandkumar},\n  year    = {2023},\n  journal = {arXiv preprint arXiv: Arxiv-2305.16291}\n}\n```\n\nDisclaimer: This project is strictly for research purposes, and not an official product from NVIDIA.\n","Voyager 是一个基于大型语言模型的开放性具身代理，能够在Minecraft中自主探索、学习技能并作出新发现。其核心功能包括自动化的探索课程设计、不断增长的行为代码库以及一种新的迭代提示机制，该机制结合了环境反馈、执行错误和自我验证来改进程序。技术上，Voyager通过与GPT-4进行黑盒查询交互，无需对模型参数进行微调即可实现复杂任务的学习。它特别适合需要持续学习能力和适应新挑战的应用场景，如游戏AI研究、教育软件开发等领域。实验表明，Voyager在获取独特物品数量、旅行距离及解锁关键技术里程碑方面均显著优于现有技术水平。",2,"2026-06-11 03:36:40","high_star"]