Skip to content

Commit 18c1edb

Browse files
authored
Update readme en (#49)
* Update README.md * update futurex * Update README.md * Update README.md * Update README.md * Update README.md * Update README.md * update * benchmark results png * Update README.md * Update README.md * Update README.md * Update README.md * Update README.md * Update README.md * Update README.md * Update README.md
1 parent 7b8771b commit 18c1edb

File tree

4 files changed

+115
-91
lines changed

4 files changed

+115
-91
lines changed

README.md

Lines changed: 108 additions & 87 deletions
Original file line numberDiff line numberDiff line change
@@ -1,18 +1,19 @@
11
<div align="center">
2-
<img src="docs/mkdocs/docs/assets/miroflow_logo.png" width="100%" alt="MiroFlow" />
2+
<img src="docs/mkdocs/docs/assets/miroflow_logo.png" width="45%" alt="MiroFlow" />
33
</div>
44

55
<br>
66

7+
78
<div align="center">
89

910
[![DOCS](https://img.shields.io/badge/Documentation-4285F4?style=for-the-badge&logo=gitbook&logoColor=white)](https://miromindai.github.io/MiroFlow/)
1011
[![DEMO](https://img.shields.io/badge/Demo-FFB300?style=for-the-badge&logo=airplayvideo&logoColor=white)](https://dr.miromind.ai/)
1112
[![MODELS](https://img.shields.io/badge/Models-5EDDD2?style=for-the-badge&logo=huggingface&logoColor=ffffff&labelColor)](https://huggingface.co/collections/miromind-ai/mirothinker-v02-68af084a18035f57b17cd902)
1213
[![DATA](https://img.shields.io/badge/Data-0040A1?style=for-the-badge&logo=huggingface&logoColor=ffffff&labelColor)](https://huggingface.co/datasets/miromind-ai/MiroVerse-v0.1)
1314

15+
[![BLOG](https://img.shields.io/badge/Blog-4285F4?style=for-the-badge&logo=google-chrome&logoColor=white)](https://miromind.ai/blog/miroflow)
1416
[![GITHUB](https://img.shields.io/badge/Github-24292F?style=for-the-badge&logo=github&logoColor=white)](https://github.com/MiroMindAI)
15-
[![WEBSITE](https://img.shields.io/badge/Website-4285F4?style=for-the-badge&logo=google-chrome&logoColor=white)](https://miromind.ai/)
1617
[![DISCORD](https://img.shields.io/badge/Discord-5865F2?style=for-the-badge&logo=discord&logoColor=white)](https://discord.com/invite/GPqEnkzQZd)
1718
[![WeChat](https://img.shields.io/badge/WeChat-07C160?style=for-the-badge&logo=wechat&logoColor=white)](https://huggingface.co/datasets/miromind-ai/MiroFlow-Benchmarks/resolve/main/assets/wechat.png)
1819
[![RedNote](https://img.shields.io/badge/RedNote-FF2442?style=for-the-badge&logo=revoltdotchat&logoColor=white)](https://www.xiaohongshu.com/user/profile/5e353bd80000000001000239)
@@ -21,96 +22,59 @@
2122

2223
<div align="center">
2324

24-
### 🚀 [Try our Demo!](https://dr.miromind.ai/) | 📚 [Full Documentation](https://miromindai.github.io/MiroFlow/)[中文](README_zh.md)[日本語](README_ja.md)
25+
### 🚀 [Try our Demo!](https://dr.miromind.ai/)[中文](README_zh.md)[日本語](README_ja.md)
2526

2627
</div>
2728

28-
<table align="center" style="border: 1px solid #ccc; border-radius: 8px; padding: 12px; background-color: #f9f9f9; width: 60%;">
29-
<tr>
30-
<td style="text-align: center; padding: 10px;">
31-
<strong>Research Assistant Demo</strong> -
32-
<span style="font-size: 0.9em; color: #555;">Read CVPR 2025 Best Paper and Provide Research Advice</span>
33-
<br>
34-
<video src="https://github.com/user-attachments/assets/99ed3172-6e9a-467a-9ccb-be45957fe2e4"
35-
controls muted preload="metadata"
36-
width="50%" height="50%"
37-
</video>
38-
</td>
39-
</tr>
40-
</table>
41-
42-
43-
## 📋 Table of Contents
44-
45-
- [📰 News & Updates](#-news--updates)
46-
- [🤖 What is MiroFlow?](#-what-is-miroflow)
47-
- [✨ Performance on Benchmarks](#-performance-on-benchmarks)
48-
- [🚀 Get Started in Under 5 Minutes](#-get-started-in-under-5-minutes)
49-
- [🤖 MiroFlow Framework](#-miroflow-ai-agentic-foundation-framework)
50-
- [🤝 Contributing](#-contributing)
51-
- [❓ FAQ](#-faq)
52-
- [📄 License & Support](#-license--support)
53-
- [👥 Acknowledgments](#-acknowledgments-and-contributors)
54-
55-
56-
## 📰 News & Updates
57-
58-
- **[2025-09-15]**: 🎉🎉 **MiroFlow v0.3** - Enhanced codebase architecture and significantly improved benchmark performance. MiroFlow now ranks #1 in the future prediction benchmark.
59-
- **[2025-08-27]**: **MiroFlow v0.2** - Achieves state-of-the-art performance across [multiple agentic benchmarks](https://miromind.ai/blog/miroflow), including HLE (27.2%), HLE-Text-Only (29.5%), BrowserComp-EN (33.2%), BrowserComp-ZH (47.1%), and xBench-DeepSearch (72.0%)
60-
- **[2025-08-26]**: Released [GAIA Validation Trace](docs/public_trace.md) (73.94% pass@1) and [Gradio Demo](https://github.com/MiroMindAI/MiroThinker/tree/main/apps/gradio-demo) for local deployment
61-
- **[2025-08-08]**: 🎉 **MiroFlow v0.1** - Complete open-source release of framework, models, and training data
29+
<div align="center">
30+
<img width="100%" alt="image" src="docs/mkdocs/docs/assets/futurex-09-12.png" />
31+
</div>
6232

6333
---
6434

65-
## 🤖 What is MiroFlow?
66-
67-
**MiroFlow** is a comprehensive framework for building intelligent AI agents that achieve state-of-the-art performance on complex reasoning tasks. It provides enhanced conversation management, flexible tool integration, and extensive benchmark evaluations across multiple datasets.
68-
69-
**MiroThinker** is the open-source agentic model series built on this framework.
35+
This 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:
7036

71-
### 🌟 Key Highlights
37+
- 🤖 **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.
38+
- 🤔 **MiroThinker**: an open-source agent foundation model that natively supports tool-assisted reasoning. See [MiroThinker](https://github.com/MiroMindAI/mirothinker).
39+
- 📊 **MiroVerse**: 147k premium open-source training data supporting research agent training. See [MiroThinker](https://huggingface.co/datasets/miromind-ai/MiroVerse-v0.1).
40+
- 🚧 **MiroTrain / MiroRL**: The training infra that supports stable and efficient training for the research agent models. See [MiroTrain](https://github.com/MiroMindAI/MiroTrain) / [MiroRL](https://github.com/MiroMindAI/MiroRL)
7241

73-
- 🏆 **State-of-the-Art Performance**: #1 ranking across [multiple agentic benchmarks](https://miromindai.github.io/MiroFlow/evaluation_overview/)
74-
- 📊 **Premium Training Data**: Curated datasets via [MiroVerse](https://huggingface.co/datasets/miromind-ai/MiroVerse-v0.1)
75-
- 🤖 **Open Models**: Complete collection at [MiroThinker](https://huggingface.co/collections/miromind-ai/mirothinker-v01-689301b6d0563321862d44a1)
76-
- 🔧 **Full Training Stack**: SFT/DPO recipes at [MiroTrain](https://github.com/MiroMindAI/MiroTrain)
77-
- 🎯 **Advanced RL**: Reinforcement learning via [MiroRL](https://github.com/MiroMindAI/MiroRL)
78-
79-
80-
### ✨ Performance on Benchmarks
81-
82-
<img width="100%" alt="image" src="docs/mkdocs/docs/assets/futurex-09-12.png" />
42+
---
8343

84-
We achieved the #1 ranking on the FutureX Benchmark Leaderboard as of September 10, 2025.
44+
## 📋 Table of Contents
8545

86-
<div align="center">
87-
<img src="docs/mkdocs/docs/assets/miroflow-0.2-performance_short.png" width="90%" alt="Comprehensive Benchmark Performance Comparison" style="border-radius: 8px; box-shadow: 0 4px 8px rgba(3, 3, 3, 0.1);">
88-
</div>
46+
- 📰 [News & Updates](#-news--updates)
47+
- 🚀 [Get Started in Under 5 Minutes](#-get-started-in-under-5-minutes)
48+
- 🤖 [What is MiroFlow?](#-what-is-miroflow)
49+
- 🌟 [Highlights](#-Highlights)
50+
-[Performance on Benchmarks](#-performance-on-benchmarks)
51+
- 🔧 [Supported Models & Tools](#-supported-models--tools)
52+
- 🤝 [Contributing](#-contributing)
53+
-[FAQ](#-faq)
54+
- 📄 [License & Support](#-license--support)
55+
- 👥 [Acknowledgments](#-acknowledgments-and-contributors)
8956

90-
We benchmark MiroFlow on a series of benchmarks including **GAIA**, **HLE**, **BrowseComp** and **xBench-DeepSearch** and achieved SOTA results.
57+
---
9158

92-
| Model/Framework | GAIA Val | HLE | HLE-Text | BrowserComp-EN | BrowserComp-ZH | xBench-DeepSearch |
93-
|----------------|----------|-----|----------|----------------|----------------|-------------------|
94-
| **MiroFlow** | **82.4%** | **27.2%** | 29.5% | 33.2% | **47.1%** | **72.0%** |
95-
| OpenAI Deep Research | 67.4% | 26.6% | - | **51.5%** | 42.9% | - |
96-
| Gemini Deep Research | - | 26.9% | - | - | - | 50+% |
97-
| Kimi Researcher | - | - | 26.9% | - | - | 69.0% |
98-
| WebSailor-72B | 55.4% | - | - | - | 30.1% | 55.0% |
99-
| Manus | 73.3% | - | - | - | - | - |
100-
| DeepSeek v3.1 | - | - | **29.8%** | - | - | 71.2% |
59+
## 📰 News & Updates
10160

61+
- **[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%.
62+
MiroFlow now ranks #1 in the future prediction benchmark. See [FutureX](https://futurex-ai.github.io/).
63+
- **[2025-08-27]**: **MiroFlow v0.2**: Achieves state-of-the-art performance across [multiple agentic benchmarks](https://miromind.ai/blog/miroflow), including HLE (27.2%), HLE-Text-Only (29.5%), BrowserComp-EN (33.2%), BrowserComp-ZH (47.1%), and xBench-DeepSearch (72.0%).
64+
- **[2025-08-26]**: Released [GAIA Validation Trace](docs/public_trace.md) (73.94% pass@1) and [Gradio Demo](https://github.com/MiroMindAI/MiroThinker/tree/main/apps/gradio-demo) for local deployment.
65+
- **[2025-08-08]**: **MiroFlow v0.1**: Complete open-source release of framework.
10266

103-
# 🚀 Get Started in Under 5 Minutes
67+
---
10468

105-
Clone the repository, configure your API key, and run your first intelligent agent. You'll just need one `OPENROUTER_API_KEY`.
69+
## 🚀 Get Started in Under 5 Minutes
10670

107-
## 📋 Prerequisites
71+
### 📋 Prerequisites
10872

10973
- **Python**: 3.12 or higher
11074
- **Package Manager**: [`uv`](https://docs.astral.sh/uv/)
11175
- **Operating System**: Linux, macOS
11276

113-
## ⚡ Quick Setup
77+
### ⚡ Quick Setup
11478

11579
**Example**: Intelligent document analysis with file processing capabilities.
11680

@@ -131,30 +95,86 @@ uv run main.py trace --config_file_name=agent_quickstart_1 --task="What is the f
13195

13296
> **💡 Tip:** If you encounter issues, check that your API key is correctly set in the `.env` file and that all dependencies are installed.
13397
134-
**🎯 Comprehensive Benchmark Suite**:
135-
- **GAIA Validation**: A benchmark for General AI Assistants. ([paper](https://arxiv.org/abs/2311.12983))
136-
- **GAIA-Text-103**: A subset of GAIA Validation for text-only tasks. ([paper](https://arxiv.org/abs/2505.22648))
137-
- **HLE**: Humanity's Last Exam. ([paper](https://arxiv.org/abs/2501.14249))
138-
- **HLE-Text-500**: A subset of HLE for text-only tasks. ([paper](https://arxiv.org/pdf/2504.21776))
139-
140-
Follow our detailed guides to reproduce benchmark results in our [Benchmarks Documentation](https://miromindai.github.io/MiroFlow/evaluation_overview/)
98+
---
14199

142-
# 🤖 MiroFlow: AI Agentic Foundation Framework
100+
## 🤖 What is MiroFlow?
143101

144-
MiroFlow is a high-performance, modular framework for building intelligent AI agents that deliver state-of-the-art results on complex reasoning tasks. 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 [workflow architecture](https://miromindai.github.io/MiroFlow/core_concepts/).
102+
MiroFlow 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://miromindai.github.io/MiroFlow/core_concepts/).
145103

146104
<div align="center">
147105
<img src="docs/mkdocs/docs/assets/miroflow_architecture.png" width="100%" alt="MiroFlow Architecture">
148106
</div>
149107

108+
<table align="center" style="border: 1px solid #ccc; border-radius: 8px; padding: 12px; background-color: #f9f9f9; width: 60%;">
109+
<tr>
110+
<td style="text-align: center; padding: 10px;">
111+
<strong>Research Assistant Demo</strong> -
112+
<span style="font-size: 0.9em; color: #555;">Read CVPR 2025 Best Paper and Provide Research Advice</span>
113+
<br>
114+
<video src="https://github.com/user-attachments/assets/99ed3172-6e9a-467a-9ccb-be45957fe2e4"
115+
controls muted preload="metadata"
116+
width="50%" height="50%"
117+
</video>
118+
</td>
119+
</tr>
120+
</table>
121+
122+
---
123+
124+
## 🌟 Highlights
125+
126+
- **Reproducible State-of-the-Art Performance**: #1 ranking across [multiple agentic benchmarks](https://miromindai.github.io/MiroFlow/evaluation_overview/), including FutureX, GAIA, HLE, xBench-DeepSearch, and BrowserComp benchmarks)
127+
- **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.
128+
- **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.
129+
130+
---
131+
132+
## 🔧 Supported Models & Tools
133+
134+
- **Models**: GPT, Claude, Gemini, Qwen, MiroThinker
135+
- **Tools**: [Audio Transcription](https://github.com/MiroMindAI/MiroFlow/blob/miroflow-v0.3/src/tool/mcp_servers/audio_mcp_server.py), [Playwright](https://github.com/MiroMindAI/MiroFlow/blob/miroflow-v0.3/src/tool/mcp_servers/browser_session.py), [Python](https://github.com/MiroMindAI/MiroFlow/blob/miroflow-v0.3/src/tool/mcp_servers/python_server.py), [File Reading](https://github.com/MiroMindAI/MiroFlow/blob/miroflow-v0.3/src/tool/mcp_servers/reading_mcp_server.py), [Reasoning](https://github.com/MiroMindAI/MiroFlow/blob/miroflow-v0.3/src/tool/mcp_servers/reasoning_mcp_server.py), [Google Search](https://github.com/MiroMindAI/MiroFlow/blob/miroflow-v0.3/src/tool/mcp_servers/searching_mcp_server.py), [VQA](https://github.com/MiroMindAI/MiroFlow/blob/miroflow-v0.3/src/tool/mcp_servers/vision_mcp_server.py), E2B
136+
137+
138+
---
139+
140+
## ✨ Performance on Benchmarks
141+
142+
We 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%.
143+
144+
<div align="center">
145+
<img width="100%" alt="image" src="docs/mkdocs/docs/assets/futurex-09-12.png" />
146+
</div>
147+
148+
We benchmark MiroFlow on a series of benchmarks, including **GAIA**, **HLE**, **BrowseComp**, and **xBench-DeepSearch**, and achieved SOTA results.
149+
150+
<img width="100%" alt="image" src="docs/mkdocs/docs/assets/benchmark_results.png" />
151+
152+
| Model/Framework | GAIA Val | HLE | HLE-Text | BrowserComp-EN | BrowserComp-ZH | xBench-DeepSearch |
153+
|----------------|----------|-----|----------|----------------|----------------|-------------------|
154+
| **MiroFlow** | **82.4%** | **27.2%** | 29.5% | 33.2% | **47.1%** | **72.0%** |
155+
| OpenAI Deep Research | 67.4% | 26.6% | - | **51.5%** | 42.9% | - |
156+
| Gemini Deep Research | - | 26.9% | - | - | - | 50+% |
157+
| Kimi Researcher | - | - | 26.9% | - | - | 69.0% |
158+
| WebSailor-72B | 55.4% | - | - | - | 30.1% | 55.0% |
159+
| Manus | 73.3% | - | - | - | - | - |
160+
| DeepSeek v3.1 | - | - | **29.8%** | - | - | 71.2% |
161+
162+
Follow our detailed guides to reproduce benchmark results in our [Benchmarks Documentation](https://miromindai.github.io/MiroFlow/evaluation_overview/)
163+
164+
---
165+
150166
## 🤝 Contributing
151167

152168
We welcome contributions from the community! Whether you're fixing bugs, adding features, or improving documentation, your help is appreciated.
153169

154170
- 📋 **Issues**: Report bugs or request features via [GitHub Issues](https://github.com/MiroMindAI/MiroFlow/issues)
171+
155172
- 🔀 **Pull Requests**: Submit improvements via pull requests
173+
156174
- 💬 **Discussions**: Join our [Discord community](https://discord.com/invite/GPqEnkzQZd) for questions and discussions
157175

176+
---
177+
158178
## ❓ FAQ
159179

160180
<details>
@@ -181,6 +201,8 @@ Follow our detailed <a href="https://miromindai.github.io/MiroFlow/evaluation_ov
181201
For commercial inquiries and enterprise support, please contact us through our <a href="https://miromind.ai/">website</a>.
182202
</details>
183203

204+
---
205+
184206
## 📄 License & Support
185207

186208
This project is licensed under the Apache License 2.0.
@@ -190,25 +212,24 @@ This project is licensed under the Apache License 2.0.
190212
<img src="https://api.star-history.com/svg?repos=MiroMindAI/MiroFlow&type=Date" alt="Star History Chart" height="300">
191213
</div>
192214

193-
### References
215+
## References
194216

195-
Technical report is coming soon!
217+
The technical report is coming soon!
196218

197219
```
198220
@misc{2025mirothinker,
199-
title={MiroFlow: An Open-Source Agentic Framework for Deep Research},
221+
title={MiroFlow: A High-Performance Open-Source Research Agent Framework},
200222
author={MiroMind AI Team},
201223
howpublished={\url{https://github.com/MiroMindAI/MiroFlow}},
202224
year={2025}
203225
}
204226
```
205227

206-
207-
208228
## 👥 Acknowledgments and Contributors
209229

210-
- **Benchmark Contributors** for the comprehensive evaluation datasets
211-
- **Open Source Community** for the tools and libraries that make this possible
230+
**Benchmark Contributors** for the comprehensive evaluation datasets
231+
232+
**Open Source Community** for the tools and libraries that make this possible
212233

213234
We thank all contributors who have helped make MiroFlow better:
214235

Lines changed: 3 additions & 0 deletions
Loading
Lines changed: 2 additions & 2 deletions
Loading
Lines changed: 2 additions & 2 deletions
Loading

0 commit comments

Comments
 (0)