[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72368":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":38,"readmeContent":39,"aiSummary":40,"trendingCount":16,"starSnapshotCount":16,"syncStatus":41,"lastSyncTime":42,"discoverSource":43},72368,"MiroFlow","MiroMindAI\u002FMiroFlow","MiroMindAI","🏆 Top-1 on 5+ benchmarks | Web UI | Supports MiroThinker, Claude, Kimi, OpenAI","https:\u002F\u002Fmiromindai.github.io\u002FMiroFlow\u002F",null,"Python",3032,313,96,1,0,9,32,65,27,99.99,"Apache License 2.0",false,"main",true,[27,28,29,30,31,32,33,34,35,36,37],"agent-framework","agents","browsecomp","claude","deep-research","futurex","gaia","gpt-5","hle","research-agent","xbench","2026-06-12 04:01:04","\u003Cdiv align=\"center\">\n  \u003Cimg src=\"docs\u002Fmkdocs\u002Fdocs\u002Fassets\u002Fmiroflow_logo.png\" width=\"45%\" alt=\"MiroFlow\" \u002F>\n\u003C\u002Fdiv>\n\n\u003Cbr> \n\n\n\u003Cdiv align=\"center\">\n\n[![DEMO](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDemo-FFB300?style=for-the-badge&logo=airplayvideo&logoColor=white)](https:\u002F\u002Fdr.miromind.ai\u002F)\n[![DATA](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FData-0040A1?style=for-the-badge&logo=huggingface&logoColor=ffffff&labelColor)](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fmiromind-ai\u002FMiroVerse-v0.1)\n[![BLOG](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBlog-4285F4?style=for-the-badge&logo=google-chrome&logoColor=white)](https:\u002F\u002Fmiromind.ai\u002Fblog\u002Fmiroflow)\n\n[![GITHUB](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FGithub-24292F?style=for-the-badge&logo=github&logoColor=white)](https:\u002F\u002Fgithub.com\u002FMiroMindAI)\n[![DISCORD](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-5865F2?style=for-the-badge&logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.com\u002Finvite\u002FGPqEnkzQZd)\n\n\u003C\u002Fdiv>\n\n\u003Cdiv align=\"center\">\n\n### 🚀 [Try Demo](https:\u002F\u002Fdr.miromind.ai\u002F) ｜ [中文](README_zh.md) ｜ [日本語](README_ja.md)\n\n\u003C\u002Fdiv>\n\n\u003Cdiv align=\"center\">\n  \u003Cimg width=\"100%\" alt=\"image\" src=\"docs\u002Fmkdocs\u002Fdocs\u002Fassets\u002Ffuturex-09-12.png\" \u002F>\n\u003C\u002Fdiv>\n\n---\n\nThis repo is the official implementation of the **MiroMind Research Agent Project**. It is a leading-performance, fully open-source system designed to perform multi-step internet research for addressing complex challenges such as future event prediction. The project currently comprises four key components:\n\n- 🤖 **MiroFlow**: an open-source research agent framework that offers reproducible state-of-the-art performance on representative benchmarks (e.g., FutureX, GAIA, HLE, xBench-DeepSearch, and BrowserComp benchmarks), included in this repo. See [[Get Started in Under 5 Minutes]](#-get-started-in-under-5-minutes) for a quick start.\n- 🤔 **MiroThinker**: an open-source agent that natively supports tool-assisted reasoning. See [MiroThinker](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002Fmirothinker).\n- 📊 **MiroVerse**: 147k premium open-source training data supporting research agent training. See [MiroVerse](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fmiromind-ai\u002FMiroVerse-v0.1).\n\n---\n\n## 📋 Table of Contents\n\n- 📰 [News & Updates](#-news--updates)\n- 🚀 [Get Started in Under 5 Minutes](#-get-started-in-under-5-minutes)\n- 🤖 [What is MiroFlow?](#-what-is-miroflow)\n- 🌟 [Highlights](#-Highlights)\n- 📈 [Performance on Benchmarks](#-performance-on-benchmarks)\n- 🔧 [Supported Models & Tools](#-supported-models--tools)\n- ❓ [FAQ](#-faq)\n- 🤝 [Contributing](#-contributing)\n- 📄 [License](#-license)\n- 🙏 [Acknowledgments](#-acknowledgments-and-contributors)\n\n---\n\n## 📰 News & Updates\n\n- **[2025-09-15]**: 🎉🎉 **MiroFlow v0.3**: Enhanced codebase architecture and significantly improved benchmark performance, boosting GPT-5's prediction accuracy for future events by 11%.\n MiroFlow now ranks #1 in the future prediction benchmark. See [FutureX](https:\u002F\u002Ffuturex-ai.github.io\u002F).\n- **[2025-08-27]**: **MiroFlow v0.2**: Achieves state-of-the-art performance across [multiple agentic benchmarks](https:\u002F\u002Fmiromind.ai\u002Fblog\u002Fmiroflow), including HLE (27.2%), HLE-Text-Only (29.5%), BrowserComp-EN (33.2%), BrowserComp-ZH (47.1%), and xBench-DeepSearch (72.0%).\n- **[2025-08-26]**: Released [GAIA Validation Trace](docs\u002Fpublic_trace.md) (73.94% pass@1) and [Gradio Demo](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroThinker\u002Ftree\u002Fmain\u002Fapps\u002Fgradio-demo) for local deployment.\n- **[2025-08-08]**: **MiroFlow v0.1**: Complete open-source release of the research agent framework.\n\n---\n\n## 🚀 Get Started in Under 5 Minutes\n\n### 📋 Prerequisites\n\n- **Python**: 3.12 or higher\n- **Package Manager**: [`uv`](https:\u002F\u002Fdocs.astral.sh\u002Fuv\u002F)\n- **Operating System**: Linux, macOS\n\n### ⚡ Quick Setup\n\n**Example**: Intelligent document analysis with file processing capabilities.\n\n```bash\n# 1. Clone and setup\ngit clone https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow && cd MiroFlow\nuv sync\n\n# 2. Configure API key\ncp .env.template .env\n# Edit .env and add your OPENROUTER_API_KEY\n\n# 3. Run your first agent\nuv run main.py trace --config_file_name=agent_quickstart_reading --task=\"What is the first country listed in the XLSX file that have names starting with Co?\" --task_file_name=\"data\u002FFSI-2023-DOWNLOAD.xlsx\"\n```\n\n🎉 **Expected Output:** Your agent should return **\\boxed{Congo Democratic Republic}** 😊\n\n> **💡 Tip:** If you encounter issues, check that your API key is correctly set in the `.env` file and that all dependencies are installed.\n\n---\n\n## 🤖 What is MiroFlow?\n\nMiroFlow is a high-performance, modular framework for building intelligent AI agents that deliver state-of-the-art results on complex reasoning tasks like future event prediction. The framework features advanced multi-turn conversation capabilities, extensive tool ecosystem integration, and hierarchical sub-agent orchestration for optimal task completion. Learn more about our [agent framework](https:\u002F\u002Fmiromindai.github.io\u002FMiroFlow\u002Fcore_concepts\u002F).\n\n\u003Cdiv align=\"center\">\n\u003Cimg src=\"docs\u002Fmkdocs\u002Fdocs\u002Fassets\u002Fmiroflow_architecture.png\" width=\"100%\" alt=\"MiroFlow Architecture\">\n\u003C\u002Fdiv>\n\n\u003Ctable align=\"center\" style=\"border: 1px solid #ccc; border-radius: 8px; padding: 12px; background-color: #f9f9f9; width: 60%;\">\n  \u003Ctr>\n    \u003Ctd style=\"text-align: center; padding: 10px;\">\n      \u003Cstrong>Research Assistant Demo\u003C\u002Fstrong> - \n      \u003Cspan style=\"font-size: 0.9em; color: #555;\">Read CVPR 2025 Best Paper and Provide Research Advice\u003C\u002Fspan>\n      \u003Cbr>\n      \u003Cvideo src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F99ed3172-6e9a-467a-9ccb-be45957fe2e4\"\n             controls muted preload=\"metadata\"\n             width=\"50%\" height=\"50%\"\n      \u003C\u002Fvideo>\n    \u003C\u002Ftd>\n  \u003C\u002Ftr>\n\u003C\u002Ftable>\n\n---\n\n## 🌟 Highlights\n\n- **Reproducible State-of-the-Art Performance**: #1 ranking across [multiple representative agentic benchmarks](https:\u002F\u002Fmiromindai.github.io\u002FMiroFlow\u002Fevaluation_overview\u002F), including FutureX, GAIA, HLE, xBench-DeepSearch, and BrowserComp benchmarks)\n- **High Concurrency & Reliability**: Built with robust concurrency management and fault-tolerant design, MiroFlow efficiently handles rate-limited APIs and unstable networks, ensuring seamless trajectory collection and reliable execution of complex tasks.\n- **Cost-Effective Deployment**: Powered by the open-source MiroThinker model, MiroFlow can run a  research agent service on a single RTX 4090. The entire stack relies on free, open-source tools, making it simple to deploy, scale, and reproduce. See [MiroThinker](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002Fmirothinker).\n\n---\n\n## 🔧 Supported Models & Tools\n\n- **Models**: GPT, Claude, Gemini, Qwen, etc.\n- **Tools**: [Audio Transcription](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow\u002Fblob\u002Fmiroflow-v0.3\u002Fsrc\u002Ftool\u002Fmcp_servers\u002Faudio_mcp_server.py), [Python](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow\u002Fblob\u002Fmiroflow-v0.3\u002Fsrc\u002Ftool\u002Fmcp_servers\u002Fpython_server.py), [File Reading](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow\u002Fblob\u002Fmiroflow-v0.3\u002Fsrc\u002Ftool\u002Fmcp_servers\u002Freading_mcp_server.py), [Reasoning](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow\u002Fblob\u002Fmiroflow-v0.3\u002Fsrc\u002Ftool\u002Fmcp_servers\u002Freasoning_mcp_server.py), [Google Search](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow\u002Fblob\u002Fmiroflow-v0.3\u002Fsrc\u002Ftool\u002Fmcp_servers\u002Fsearching_mcp_server.py), [VQA](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow\u002Fblob\u002Fmiroflow-v0.3\u002Fsrc\u002Ftool\u002Fmcp_servers\u002Fvision_mcp_server.py), E2B, etc.\n\n\n---\n\n## 📈 Performance on Benchmarks\n\nWe achieved the #1 ranking on the FutureX Benchmark Leaderboard as of September 10, 2025, boosting GPT-5's prediction accuracy for future events by 11%.\n\n\u003Cdiv align=\"center\">\n  \u003Cimg width=\"100%\" alt=\"image\" src=\"docs\u002Fmkdocs\u002Fdocs\u002Fassets\u002Ffuturex-09-12.png\" \u002F>\n\u003C\u002Fdiv>\n\nWe benchmark MiroFlow on a series of benchmarks, including **GAIA**, **HLE**, **BrowseComp**, and **xBench-DeepSearch**, and achieved SOTA results.\n\n\u003Cimg width=\"100%\" alt=\"image\" src=\"docs\u002Fmkdocs\u002Fdocs\u002Fassets\u002Fbenchmark_results.png\" \u002F>\n\n| Model\u002FFramework | GAIA Val | HLE | HLE-Text | BrowserComp-EN | BrowserComp-ZH | xBench-DeepSearch |\n|----------------|----------|-----|----------|----------------|----------------|-------------------|\n| **MiroFlow** | **82.4%** | **27.2%** | 29.5% | 33.2% | **47.1%** | **72.0%** |\n| OpenAI Deep Research | 67.4% | 26.6% | - | **51.5%** | 42.9% | - |\n| Gemini Deep Research | - | 26.9% | - | - | - | 50+% |\n| Kimi Researcher | - | - | 26.9% | - | - | 69.0% |\n| WebSailor-72B | 55.4% | - | - | - | 30.1% | 55.0% |\n| Manus | 73.3% | - | - | - | - | - |\n| DeepSeek v3.1 | - | - | **29.8%** | - | - | 71.2% |\n\nFollow our detailed guides to reproduce benchmark results in our [Benchmarks Documentation](https:\u002F\u002Fmiromindai.github.io\u002FMiroFlow\u002Fevaluation_overview\u002F)\n\n---\n\n## ❓ FAQ\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>What API keys do I need?\u003C\u002Fstrong>\u003C\u002Fsummary>\n\u003Cbr>\nYou only need an OpenRouter API key to get started. OpenRouter provides access to multiple language models through a single API.\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>Can I use other language models besides OpenRouter?\u003C\u002Fstrong>\u003C\u002Fsummary>\n\u003Cbr>\nYes, MiroFlow supports various language models. Check our documentation for configuration details.\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>How do I reproduce the benchmark results?\u003C\u002Fstrong>\u003C\u002Fsummary>\n\u003Cbr>\nFollow our detailed \u003Ca href=\"https:\u002F\u002Fmiromindai.github.io\u002FMiroFlow\u002Fevaluation_overview\u002F\">Benchmarks Documentation\u003C\u002Fa> for step-by-step reproduction guides.\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>Is there commercial support available?\u003C\u002Fstrong>\u003C\u002Fsummary>\n\u003Cbr>\nFor commercial inquiries and enterprise support, please contact us through our \u003Ca href=\"https:\u002F\u002Fmiromind.ai\u002F\">website\u003C\u002Fa>.\n\u003C\u002Fdetails>\n\n---\n\n## 🤝 Contributing\n\nWe welcome contributions from the community! Whether you're fixing bugs, adding features, or improving documentation, your help is appreciated.\n\n- 📋 **Issues**: Report bugs or request features via [GitHub Issues](https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow\u002Fissues).\n- 🔀 **Pull Requests**: Submit improvements via pull requests.\n- 💬 **Discussions**: Join our [Discord community](https:\u002F\u002Fdiscord.com\u002Finvite\u002FGPqEnkzQZd) for questions and discussions.\n\n\n## 📄 License\n\nThis project is licensed under the Apache License 2.0.\n\n## 🙏 Acknowledgments\n\n**Benchmark Contributors** for the comprehensive evaluation datasets.\n\n**Open Source Community** for the tools and libraries that make this possible.\n\nWe thank all contributors who have helped make MiroFlow better:\n\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FMiroMindAI\u002FMiroFlow\u002Fgraphs\u002Fcontributors\">\n  \u003Cimg src=\"https:\u002F\u002Fcontrib.rocks\u002Fimage?repo=MiroMindAI\u002FMiroFlow\" \u002F>\n\u003C\u002Fa>\n\nJoin our community and help us build the future of AI agents!\n\n## References\n\n```\n@article{miromind2025mirothinker,\n  title={MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scaling},\n  author={MiroMind Team and Bai, Song and Bing, Lidong and Chen, Carson and Chen, Guanzheng and Chen, Yuntao and Chen, Zhe and Chen, Ziyi and Dong, Xuan and others},\n  journal={arXiv preprint arXiv:2511.11793},\n  year={2025}\n}\n```\n\n","MiroFlow 是一个高性能的开源研究代理框架，旨在通过多步骤互联网研究解决复杂的挑战，如未来事件预测。该项目的核心功能包括支持多种模型（如MiroThinker、Claude、Kimi和OpenAI）以及提供了一个用户友好的Web界面。技术上，MiroFlow 在多个基准测试中表现出色，并且其架构设计易于扩展与定制。此外，它还配备了147,000条高质量训练数据集 MiroVerse，以支持更深入的研究。适用于需要进行深度网络搜索分析、预测未来趋势或利用人工智能辅助决策的各种场景。",2,"2026-06-11 03:41:32","high_star"]