[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72612":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":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":27,"readmeContent":28,"aiSummary":29,"trendingCount":16,"starSnapshotCount":16,"syncStatus":30,"lastSyncTime":31,"discoverSource":32},72612,"dgm","jennyzzt\u002Fdgm","jennyzzt","Darwin Gödel Machine: Open-Ended Evolution of Self-Improving Agents","",null,"Python",2097,420,51,16,0,9,22,59,27,94.77,"Apache License 2.0",false,"main",true,[],"2026-06-12 04:01:06","\u003Ch1 align=\"center\">\n    Darwin Gödel Machine:\u003Cbr\u002F>Open-Ended Evolution of Self-Improving Agents\n\u003C\u002Fh1>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fjennyzzt\u002Fdgm\u002Fblob\u002Fmain\u002FLICENSE\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache%202.0-blue.svg?style=for-the-badge\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.22954\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv-2505.22954-b31b1b.svg?logo=arxiv&style=for-the-badge\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fsakana.ai\u002Fdgm\u002F\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Blog-%238D6748?style=for-the-badge&logo=Website&logoColor=white\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fx.com\u002FSakanaAILabs\u002Fstatus\u002F1928272612431646943\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Ftwitter-%230077B5.svg?&style=for-the-badge&logo=twitter&logoColor=white&color=00acee\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F1Kcu9TbIa9Z50pJ7S6hH9omzzD1pxIYZC?usp=sharing\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FExperiment%20Logs-4285F4?style=for-the-badge&logo=googledrive&logoColor=white\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n\nRepository for **Darwin Gödel Machine (DGM)**, a novel self-improving system that iteratively modifies its own code (thereby also improving its ability to modify its own codebase) and empirically validates each change using coding benchmarks.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\".\u002Fmisc\u002Foverview.gif\" width=\"100%\" height=\"auto\" \u002F>\n\u003C\u002Fp>\n\u003C!-- \u003Cp align=\"center\">\n\u003Cimg src=\".\u002Fmisc\u002Fconceptual.svg\"\u002F>\u003C\u002Fa>\u003Cbr>\n\u003C\u002Fp> -->\n\n\n## Setup\n```bash\n# API keys, add to ~\u002F.bashrc\nexport OPENAI_API_KEY='...'\nexport ANTHROPIC_API_KEY='...'\n```\n\n```bash\n# Verify that Docker is properly configured in your environment.\ndocker run hello-world\n \n# If a permission error occurs, add the user to the Docker group\nsudo usermod -aG docker $USER\nnewgrp docker\n```\n\n```bash\n# Install dependencies\npython3 -m venv venv\nsource venv\u002Fbin\u002Factivate\npip install -r requirements.txt\n\n# Optional: for running analysis\nsudo apt-get install graphviz graphviz-dev\npip install -r requirements_dev.txt\n```\n\n```bash\n# Clone SWE-bench\ncd swe_bench\ngit clone https:\u002F\u002Fgithub.com\u002Fprinceton-nlp\u002FSWE-bench.git\ncd SWE-bench\ngit checkout dc4c087c2b9e4cefebf2e3d201d27e36\npip install -e .\ncd ..\u002F..\u002F\n\n# Prepare Polyglot\n# Make sure git is properly configured in your environment with username and email\npython -m polyglot.prepare_polyglot_dataset\n```\n\n## Running the DGM\n```bash\npython DGM_outer.py\n```\nBy default, outputs will be saved in the `output_dgm\u002F` directory.\n\n## File Structure\n- `analysis\u002F` scripts used for plotting and analysis\n- `initial\u002F` SWE-bench logs and performance of the initial agent\n- `initial_polyglot\u002F` Polyglot logs and performance of the initial agent\n- `swe_bench\u002F` code needed for SWE-bench evaluation\n- `polyglot\u002F` code needed for Polyglot evaluation\n- `prompts\u002F` prompts used for foundation models\n- `tests\u002F` tests for the DGM system\n- `tools\u002F` tools available to the foundation models\n- `coding_agent.py` main implementation of the initial coding agent\n- `DGM_outer.py` entry point for running the DGM algorithm\n\n## Logs from Experiments\nThis [google drive folder](https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F1Kcu9TbIa9Z50pJ7S6hH9omzzD1pxIYZC?usp=sharing) contains all the foundation model output logs from the experiments shown in the paper.\n\n## Safety Consideration\n> [!WARNING]  \n> This repository involves executing untrusted, model-generated code. We strongly advise users to be aware of the associated safety risks. While it is highly unlikely that such code will perform overtly malicious actions under our current settings and with the models we use, it may still behave destructively due to limitations in model capability or alignment. By using this repository, you acknowledge and accept these risks.\n\n## Acknowledgement\n\nThe evaluation framework implementations are based on the [SWE-bench](https:\u002F\u002Fgithub.com\u002Fswe-bench\u002FSWE-bench) and [polyglot-benchmark](https:\u002F\u002Fgithub.com\u002FAider-AI\u002Fpolyglot-benchmark) repositories.\n\n## Citing\nIf you find this project useful, please consider citing:\n```bibtex\n@article{zhang2025darwin,\n  title={Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents},\n  author={Zhang, Jenny and Hu, Shengran and Lu, Cong and Lange, Robert and Clune, Jeff},\n  journal={arXiv preprint arXiv:2505.22954},\n  year={2025}\n}\n```\n","Darwin Gödel Machine (DGM) 是一个能够自我改进的系统，通过迭代修改自身代码并利用编码基准进行实证验证来不断提升其性能。该项目的核心功能包括自动化的代码优化和自我学习能力，采用Python语言开发，并依赖于OpenAI和Anthropic等API提供支持。技术特点在于结合了机器学习与进化算法，使系统能够在开放式的环境中持续进化。适用于需要自适应、自动化改进程序性能的研究场景或开发项目中，特别是在软件工程和人工智能交叉领域有较高的应用价值。",2,"2026-06-11 03:42:46","high_star"]