[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72145":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":25,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":41,"readmeContent":42,"aiSummary":43,"trendingCount":16,"starSnapshotCount":16,"syncStatus":44,"lastSyncTime":45,"discoverSource":46},72145,"EverOS","EverMind-AI\u002FEverOS","EverMind-AI","Self-evolving memory across Agent and platform. The one portable memory layer for every agent they use - Claude Code, Codex, OpenClaw, Hermes, and more","https:\u002F\u002Fevermind.ai",null,"Python",7266,719,101,22,0,248,772,1678,744,114.57,"Apache License 2.0",false,"main",true,[27,28,29,30,31,32,33,34,35,36,37,38,39,40],"agent-memory","agentic-ai","ai","chats","clawdbot","clawdbot-skill","llm","long-term-memory","mcp","memory","memory-management","python3","rag","skills","2026-06-12 04:01:03","\u003Cdiv align=\"center\" id=\"readme-top\">\n\n![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F0bf97efd-580f-4a53-a2a2-58d6daea7290)\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fx.com\u002Fevermind\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEverMind-000000?labelColor=gray&style=for-the-badge&logo=x&logoColor=white\" alt=\"X\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002FEverMind-AI\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F🤗_HuggingFace-EverMind-F5C842?labelColor=gray&style=for-the-badge\" alt=\"HuggingFace\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002FgYep5nQRZJ\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fdynamic\u002Fjson?url=https%3A%2F%2Fdiscord.com%2Fapi%2Fv10%2Finvites%2FgYep5nQRZJ%3Fwith_counts%3Dtrue&query=%24.approximate_presence_count&suffix=%20online&label=Discord&color=404EED&labelColor=gray&style=for-the-badge&logo=discord&logoColor=white\" alt=\"Discord\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FEverMind-AI\u002FEverOS\u002Fdiscussions\u002F67\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeCom-EverMind_社区-07C160?labelColor=gray&style=for-the-badge&logo=wechat&logoColor=white\" alt=\"WeChat\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n[Website](https:\u002F\u002Fevermind.ai) · [Documentation](https:\u002F\u002Fdocs.evermind.ai) · [Blog](https:\u002F\u002Fevermind.ai\u002Fblogs)\n\n\u003C\u002Fdiv>\n\n\n\u003Cbr>\n\n\u003Cdetails open>\n  \u003Csummary>\u003Ckbd>Table of Contents\u003C\u002Fkbd>\u003C\u002Fsummary>\n\n\u003Cbr>\n\n- [Project Overview](#project-overview)\n- [Use Cases](#use-cases)\n- [Quick Start](#quick-start)\n- [Architecture Methods](#architecture-methods)\n- [Benchmarks](#benchmarks)\n- [Evaluation](#evaluation)\n- [Citations](#citations)\n- [Stay Tuned](#stay-tuned)\n- [Contributing](#contributing)\n\n\u003Cbr>\n\n\u003C\u002Fdetails>\n\n\n\n## Project Overview\n\n**EverOS** is a unified home for applying, building, and evaluating long-term memory in self-evolving agents. The repository is organized around three essential parts:\n\n| Part | What it gives you | Start here |\n| :--- | :--- | :--- |\n| **Use cases** | Apps, demos, and integrations showing how memory changes real agent workflows. | [use-cases\u002F](use-cases\u002F) |\n| **Architecture methods** | Memory systems and algorithms you can run, extend, or compare. | [methods\u002F](methods\u002F) |\n| **Benchmarks** | Open evaluation suites for memory quality and agent self-evolution. | [benchmarks\u002F](benchmarks\u002F) |\n\nAt the center of EverOS is **EverCore**, a long-term memory operating system for agents. If you are new to the project, scan the use cases first to see what memory enables, then follow the [Quick Start](#quick-start) to run EverCore locally. The architecture and benchmark sections below give you the deeper reference material when you are ready to compare systems or reproduce results.\n\n\u003Cbr>\n\n## Use Cases\n\nUse cases show what persistent memory makes possible in real products and workflows. Some examples are packaged in this repository; others point to external demos or integrations you can study and adapt.\n\n\u003Ctable>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F650b901b-c9ba-4001-bac7-626b009df830)\n\n#### Rokid AI Assistant with EverOS\n\nConnect to EverOS within Rokid Glasses enabling long-term memory for all of your smart activities.\n\nComing soon\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F85b338b2-e48e-4a65-9f30-0bc6998df872)\n\n#### Creative Assistant with Memory\n\nCreative assistant with long-term memory, never forget your crativites anymore.\n\nComing soon\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Ff30617a1-adc0-4271-bc0e-c3a0b28cb903)](https:\u002F\u002Fgithub.com\u002Fxunyud\u002FEarth-Online)\n\n#### Earth Online Memory Game\n\nEarth Online is a memory-aware productivity game that turns everyday planning into a living quest log.\n\n[Code](https:\u002F\u002Fgithub.com\u002Fxunyud\u002FEarth-Online)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F57d8cda7-35a5-4561-b794-5520dffc917b)](https:\u002F\u002Fgithub.com\u002Fgolutra\u002Fgolutra) \n\n#### Multi-Agent Orchestration Platform\n\nGolutra presents a multi-agent workforce for engineering teams, extending the IDE model from a single assistant to coordinated agents.\n\n[Code](https:\u002F\u002Fgithub.com\u002Fgolutra\u002Fgolutra)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F75f19db5-30f6-4eed-9b1e-c9c6a0e6b7de)](https:\u002F\u002Fgithub.com\u002FYangtze-Seventh\u002Ftaste-verse)\n\n#### Your Personal Tasting Universe\n\nRecord, visualize, and explore your tasting journey through an immersive 3D star map.\n\n[Code](https:\u002F\u002Fgithub.com\u002FYangtze-Seventh\u002Ftaste-verse)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F93ac2a68-4f18-4fcb-8d87-80aeb00a9d7c)](https:\u002F\u002Fgithub.com\u002Fkellyvv\u002FOpenHer) \n\n#### EverOS Open Her\n\nBuild AI that feels. Open-source persona engine — personality emerges from neural drives, not prompts. Inspired by Her.\n\n[Code](https:\u002F\u002Fgithub.com\u002Fkellyvv\u002FOpenHer)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F550071c1-dc39-4964-9f67-ffdfad792345)](https:\u002F\u002Fchromewebstore.google.com\u002Fdetail\u002Fruminer-browser-agent\u002Flbccjohfpdpimbhpckljimgolndfmfif)\n\n#### Browser Agent for Personal Memory\n\nRuminer brings persistent memory to a browser agent so it can carry personal context across web tasks.\n\n[Plugin](https:\u002F\u002Fchromewebstore.google.com\u002Fdetail\u002Fruminer-browser-agent\u002Flbccjohfpdpimbhpckljimgolndfmfif)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fc258a6c4-fe70-497a-98d1-3dade4a932f6)](https:\u002F\u002Fgithub.com\u002Fnanxingw\u002FEverMem) \n\n#### EverMem Sync with EverOS\n\nOne command to connect any AI coding CLI to EverMemOS long-term memory.\n\n[Code](https:\u002F\u002Fgithub.com\u002Fnanxingw\u002FEverMem)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F39274473-ceb3-48fb-a031-e22230decbe2)](https:\u002F\u002Fgithub.com\u002Fmco-org\u002Fmco)\n\n#### MCO - Orchestrate AI Coding Agents\n\nMCO equips your primary agent with an agent team that can work together to solve complex tasks.\n\n[Code](https:\u002F\u002Fgithub.com\u002Fmco-org\u002Fmco)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F314c9126-8e08-4688-bbbb-8555ad58cf67)](https:\u002F\u002Fgithub.com\u002Fonenewborn\u002FStudyBuddy-public) \n\n#### Study Buddy with Self-Evolving Memory\n\nStudy proactively with an agent that has self-evolving memory.\n\n[Code](https:\u002F\u002Fgithub.com\u002Fonenewborn\u002FStudyBuddy-public)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F21da76aa-9a8b-48e0-9134-42429d7390e7)](https:\u002F\u002Fgithub.com\u002FTonyLiangDesign\u002FMemoCare)\n\n#### Alzheimer’s Memory Assistant\n\nEmpowering individuals with advanced memory support and daily assistance.\n\n[Code](https:\u002F\u002Fgithub.com\u002FTonyLiangDesign\u002FMemoCare)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fe2428df3-ea11-4e88-8f9c-dad437dd8998)](https:\u002F\u002Fgithub.com\u002FAlexL1024\u002FNeuralConnect) \n\n#### Memory-Driven Multi-Agent NPC Experience\n\nAn iOS sci-fi mystery game where players explore and uncover the truth.\n\n[Code](https:\u002F\u002Fgithub.com\u002FAlexL1024\u002FNeuralConnect)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fe6eaf308-a874-483f-8874-6934bf95a78f)](https:\u002F\u002Fgithub.com\u002Felontusk5219-prog\u002FMobi)\n\n#### Mobi Companion\n\nAn iOS app where users create, nurture, and live with a personalized AI companion called Mobi.\n\n[Code](https:\u002F\u002Fgithub.com\u002Felontusk5219-prog\u002FMobi)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F9aabcaa9-f97a-49d2-9109-0b5bb696ed41)](https:\u002F\u002Fgithub.com\u002FJaMesLiMers\u002FEvermemCompetition-Spiro)\n\n#### AI Wearable with Memory\n\nA context-native AI wearable that listens to everyday life and converts conversations into memory.\n\n[Code](https:\u002F\u002Fgithub.com\u002FJaMesLiMers\u002FEvermemCompetition-Spiro)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fdf9677ec-386f-4c56-a428-08bca25c54dc)](https:\u002F\u002Fgithub.com\u002FEverMind-AI\u002Feveros\u002Ftree\u002Fagent_memory\u002Feveros-openclaw-plugin)\n\n#### OpenClaw Agent Memory\n\nA 24\u002F7 agent workflow with continuous learning memory across sessions.\n\n[Plugin](https:\u002F\u002Fgithub.com\u002FEverMind-AI\u002Feveros\u002Ftree\u002Fagent_memory\u002Feveros-openclaw-plugin)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F3a2357a1-c0c3-464a-8979-0d1cdfc9b0d4)](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-framework\u002Ftree\u002Fmain\u002Fai_agents\u002Fagents\u002Fexamples\u002Fvoice-assistant-with-everos)\n\n#### Live2D Character with Memory\n\nAdd long-term memory to a real-time Live2D character, powered by [TEN Framework](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-framework).\n\n[Code](https:\u002F\u002Fgithub.com\u002FTEN-framework\u002Ften-framework\u002Ftree\u002Fmain\u002Fai_agents\u002Fagents\u002Fexamples\u002Fvoice-assistant-with-everos)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fc36bdc04-97d3-4fe9-97d9-4b93b475595a)](https:\u002F\u002Fscreenshot-analysis-vercel.vercel.app\u002F)\n\n#### Computer-Use with Memory\n\nRun screenshot-based analysis with computer-use and store the results in memory.\n\n[Live Demo](https:\u002F\u002Fscreenshot-analysis-vercel.vercel.app\u002F)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F54a7cf8f-62c4-4fbc-9d50-b214d034e051)](use-cases\u002Fgame-of-throne-demo)\n\n#### Game of Thrones Memories\n\nA demonstration of AI memory infrastructure through an interactive Q&A experience with *A Game of Thrones*.\n\n[Code](use-cases\u002Fgame-of-throne-demo)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Faf37c1f6-7ba5-430c-b99d-2a7e7eac618f)](use-cases\u002Fclaude-code-plugin)\n\n#### Claude Code Plugin\n\nPersistent memory for Claude Code. Automatically saves and recalls context from past coding sessions.\n\n[Code](use-cases\u002Fclaude-code-plugin)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n[![banner-gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fd521d28c-0ccd-44ff-aecc-828245e2f973)](https:\u002F\u002Fmain.d2j21qxnymu6wl.amplifyapp.com\u002Fgraph.html)\n\n#### Memory Graph Visualization\n\nExplore stored entities and relationships in a graph interface. Frontend demo; backend integration is in progress.\n\n[Live Demo](https:\u002F\u002Fmain.d2j21qxnymu6wl.amplifyapp.com\u002Fgraph.html)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftable>\n\n\u003Cbr>\n\u003Cdiv align=\"right\">\n\n[![](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003C\u002Fdiv>\n\n## Quick Start\n\nChoose the path that matches your goal:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FEverMind-AI\u002FEverOS.git\ncd EverOS\n```\n\n| Goal | Component | Entry Point |\n| :--- | :--- | :--- |\n| Build agents with long-term memory | **EverCore** | [methods\u002FEverCore\u002F](methods\u002FEverCore\u002F) |\n| Explore the hypergraph memory architecture | **HyperMem** | [methods\u002FHyperMem\u002F](methods\u002FHyperMem\u002F) |\n| Evaluate memory system quality | **EverMemBench** | [benchmarks\u002FEverMemBench\u002F](benchmarks\u002FEverMemBench\u002F) |\n| Measure agent self-evolution | **EvoAgentBench** | [benchmarks\u002FEvoAgentBench\u002F](benchmarks\u002FEvoAgentBench\u002F) |\n| Adapt an example app or integration | **Use cases** | [use-cases\u002F](use-cases\u002F) |\n\n> Each component has its own installation guide, dependency configuration, and usage examples.\n\n### EverCore\n\nThe fastest way to run a memory system locally is to start with EverCore:\n\n```bash\ncd methods\u002FEverCore\n\n# Start Docker services\ndocker compose up -d\n\n# Install dependencies\ncurl -LsSf https:\u002F\u002Fastral.sh\u002Fuv\u002Finstall.sh | sh\nuv sync\n\n# Configure API keys\ncp env.template .env\n# Edit .env and set:\n#   - LLM_API_KEY (for memory extraction)\n#   - VECTORIZE_API_KEY (for embedding\u002Frerank)\n\n# Start server\nuv run python src\u002Frun.py\n\n# Verify installation\ncurl http:\u002F\u002Flocalhost:1995\u002Fhealth\n# Expected response: {\"status\": \"healthy\", ...}\n```\n\nServer runs at `http:\u002F\u002Flocalhost:1995` · [Full Setup Guide](methods\u002FEverCore\u002Fdocs\u002Finstallation\u002FSETUP.md)\n\n### Basic Usage\n\nStore and retrieve memories with simple Python code:\n\n```python\nimport requests\n\nAPI_BASE = \"http:\u002F\u002Flocalhost:1995\u002Fapi\u002Fv1\"\n\n# 1. Store a conversation memory\nrequests.post(f\"{API_BASE}\u002Fmemories\", json={\n    \"message_id\": \"msg_001\",\n    \"create_time\": \"2025-02-01T10:00:00+00:00\",\n    \"sender\": \"user_001\",\n    \"content\": \"I love playing soccer on weekends\"\n})\n\n# 2. Search for relevant memories\nresponse = requests.get(f\"{API_BASE}\u002Fmemories\u002Fsearch\", json={\n    \"query\": \"What sports does the user like?\",\n    \"user_id\": \"user_001\",\n    \"memory_types\": [\"episodic_memory\"],\n    \"retrieve_method\": \"hybrid\"\n})\n\nresult = response.json().get(\"result\", {})\nfor memory_group in result.get(\"memories\", []):\n    print(f\"Memory: {memory_group}\")\n```\n\n[More Examples](methods\u002FEverCore\u002Fdocs\u002Fusage\u002FUSAGE_EXAMPLES.md) · [API Reference](https:\u002F\u002Fdocs.evermind.ai\u002Fapi-reference\u002Fintroduction) · [Interactive Demos](methods\u002FEverCore\u002Fdocs\u002Fusage\u002FDEMOS.md)\n\n\u003Cbr>\n\u003Cdiv align=\"right\">\n\n[![](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003C\u002Fdiv>\n\n## Architecture Methods\n\nThese are the memory architectures currently included in EverOS. Use them as runnable systems, research references, or starting points for your own agent memory layer.\n\n\u003Ctable>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n### EverCore\n\nA self-organizing memory operating system inspired by biological imprinting. Extracts, structures, and retrieves long-term knowledge from conversations so agents can remember, understand, and continuously evolve.\n\nLoCoMo **93.05%** · LongMemEval **83.00%**\n\n[Paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.02163) · [Docs](methods\u002FEverCore\u002F)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n### HyperMem\n\nA hypergraph-based hierarchical memory architecture that captures high-order associations through hyperedges, with topic, event, and fact layers for coarse-to-fine conversation retrieval.\n\nLoCoMo **92.73%**\n\n[Paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2604.08256) · [Docs](methods\u002FHyperMem\u002F)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftable>\n\n\u003Cbr>\n\u003Cdiv align=\"right\">\n\n[![](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003C\u002Fdiv>\n\n## Benchmarks\n\nThese benchmarks provide shared standards for measuring memory quality and agent self-evolution across systems.\n\n\u003Ctable>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n### EverMemBench\n\nThree-layer memory quality evaluation: factual recall, applied reasoning, and personalized generalization.\n\n[Paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.01313) · [Dataset](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FEverMind-AI\u002FEverMemBench-Dynamic) · [Docs](benchmarks\u002FEverMemBench\u002F)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n### EvoAgentBench\n\nAgent self-evolution evaluation through longitudinal growth curves, transfer efficiency, error avoidance, and skill-hit quality.\n\n[Dataset](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FEverMind-AI\u002FEvoAgentBench) · [Docs](benchmarks\u002FEvoAgentBench\u002F)\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftable>\n\n\u003Cbr>\n\u003Cdiv align=\"right\">\n\n[![](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003C\u002Fdiv>\n\n## Evaluation\n\nUse the evaluation runner to reproduce EverCore results or compare another memory system against the same benchmark tasks.\n\n### Benchmark Results\n\n![EverOS Benchmark Results](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F41b656e7-6f82-41b7-891d-d6079d10dd39)\n\n### Supported Benchmarks\n\n- **[LoCoMo](https:\u002F\u002Fgithub.com\u002Fsnap-research\u002Flocomo)** — Long-context memory benchmark with single\u002Fmulti-hop reasoning\n- **[LongMemEval](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fxiaowu0162\u002Flongmemeval-cleaned)** — Multi-session conversation evaluation\n- **[PersonaMem](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fbowen-upenn\u002FPersonaMem)** — Persona-based memory evaluation\n\n### Run Evaluations\n\n```bash\ncd methods\u002FEverCore\n\n# Install evaluation dependencies\nuv sync --group evaluation\n\n# Run smoke test (quick verification)\nuv run python -m evaluation.cli --dataset locomo --system everos --smoke\n\n# Run full evaluation\nuv run python -m evaluation.cli --dataset locomo --system everos\n\n# View results\ncat evaluation\u002Fresults\u002Flocomo-everos\u002Freport.txt\n```\n\n[Full Evaluation Guide](methods\u002FEverCore\u002Fevaluation\u002FREADME.md) · [Complete Results](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FEverMind-AI\u002Feveros_Eval_Results)\n\n\u003Cbr>\n\u003Cdiv align=\"right\">\n\n[![](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003C\u002Fdiv>\n\n## Citations\n\nIf EverOS helps your research, please cite the relevant paper:\n\n```bibtex\n@article{hu2026evermemos,\n  title   = {EverMemOS: A Self-Organizing Memory Operating System for Structured Long-Horizon Reasoning},\n  author  = {Chuanrui Hu and Xingze Gao and Zuyi Zhou and Dannong Xu and Yi Bai and Xintong Li and Hui Zhang and Tong Li and Chong Zhang and Lidong Bing and Yafeng Deng},\n  journal = {arXiv preprint arXiv:2601.02163},\n  year    = {2026}\n}\n\n@article{yue2026hypermem,\n  title   = {HyperMem: Hypergraph Memory for Long-Term Conversations},\n  author  = {Juwei Yue and Chuanrui Hu and Jiawei Sheng and Zuyi Zhou and Wenyuan Zhang and Tingwen Liu and Li Guo and Yafeng Deng},\n  journal = {arXiv preprint arXiv:2604.08256},\n  year    = {2026}\n}\n\n@article{hu2026evaluating,\n  title   = {Evaluating Long-Horizon Memory for Multi-Party Collaborative Dialogues},\n  author  = {Chuanrui Hu and Tong Li and Xingze Gao and Hongda Chen and Yi Bai and Dannong Xu and Tianwei Lin and Xiaohong Li and Yunyun Han and Jian Pei and Yafeng Deng},\n  journal = {arXiv preprint arXiv:2602.01313},\n  year    = {2026}\n}\n```\n\n\u003Cbr>\n\u003Cdiv align=\"right\">\n\n[![](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003C\u002Fdiv>\n\n## Stay Tuned\n\nStar the repo or join the community links above to follow new architecture methods, benchmark releases, and memory-enabled use cases.\n\n![star us gif](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F0c512570-945a-483a-9f47-8e067bd34484)\n\n\u003Cbr>\n\u003Cdiv align=\"right\">\n\n[![](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003C\u002Fdiv>\n\n## Contributing\n\nContributions are welcome across the whole repository: architecture methods, benchmark coverage, use-case examples, documentation, and bug fixes. Browse [Issues](https:\u002F\u002Fgithub.com\u002FEverMind-AI\u002FEverOS\u002Fissues) to find a good entry point, then open a PR when you are ready.\n\n\u003Cbr>\n\n> [!TIP]\n>\n> **Welcome all kinds of contributions** 🎉\n>\n> Help make EverOS better. Code, documentation, benchmark reports, use-case write-ups, and integration examples are all valuable. Share your projects on social media to inspire others.\n>\n> Connect with one of the EverOS maintainers [@elliotchen200](https:\u002F\u002Fx.com\u002Felliotchen200) on 𝕏 or [@cyfyifanchen](https:\u002F\u002Fgithub.com\u002Fcyfyifanchen) on GitHub for project updates, discussions, and collaboration opportunities.\n\n![divider](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F2e2bbcc6-e6d8-4227-83c6-0620fc96f761#gh-light-mode-only)\n![divider](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fd57fad08-4f49-4a1c-bdfc-f659a5d86150#gh-dark-mode-only)\n\n### Code Contributors\n\n[![EverOS Contributors](https:\u002F\u002Fcontrib.rocks\u002Fimage?repo=EverMind-AI\u002FEverOS)](https:\u002F\u002Fgithub.com\u002FEverMind-AI\u002FEverOS\u002Fgraphs\u002Fcontributors)\n\n![divider](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F2e2bbcc6-e6d8-4227-83c6-0620fc96f761#gh-light-mode-only)\n![divider](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fd57fad08-4f49-4a1c-bdfc-f659a5d86150#gh-dark-mode-only)\n\n### Contribution Guidelines\n\nRead the [Contribution Guidelines](.github\u002FCONTRIBUTING.md) for setup, pull request expectations, and use-case submission notes. For responsible disclosure, see the [Security Policy](.github\u002FSECURITY.md).\n\n![divider](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F2e2bbcc6-e6d8-4227-83c6-0620fc96f761#gh-light-mode-only)\n![divider](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fd57fad08-4f49-4a1c-bdfc-f659a5d86150#gh-dark-mode-only)\n\n### License, Conduct, and Acknowledgments\n\n[Apache 2.0](https:\u002F\u002Fgithub.com\u002FEverMind-AI\u002FEverOS\u002Fblob\u002Fmain\u002FLICENSE) • [Code of Conduct](.github\u002FCODE_OF_CONDUCT.md) • [Acknowledgments](methods\u002FEverCore\u002Fdocs\u002FACKNOWLEDGMENTS.md)\n\n\u003Cbr>\n\n\u003Cdiv align=\"right\">\n\n[![](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003C\u002Fdiv>\n","EverOS 是一个用于构建、评估和集成自进化代理长期记忆的统一平台。其核心功能包括通过EverCore系统实现长期记忆管理，支持多种内存系统与算法的扩展及比较，并提供了开放的评估套件来测试记忆质量和代理自我进化能力。项目采用Python语言编写，具有良好的可扩展性和灵活性。它适用于需要为AI助手或其他智能体添加持久化记忆以改进用户体验或增强系统性能的各种场景，比如聊天机器人、个性化推荐等。此外，EverOS还提供了一系列实际应用案例和基准测试工具，帮助开发者快速上手并深入理解如何利用长期记忆提升代理的表现。",2,"2026-06-11 03:40:35","high_star"]