Spaces:
Sleeping
Sleeping
Update README.md
Browse files
README.md
CHANGED
@@ -1,122 +1,8 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
English | [简体中文](README_zh-CN.md)
|
10 |
-
|
11 |
-
<https://github.com/user-attachments/assets/44ffe4b9-be26-4b93-a77b-02fed16e33fe>
|
12 |
-
|
13 |
-
</div>
|
14 |
-
</p>
|
15 |
-
|
16 |
-
## ✨ MindSearch: Mimicking Human Minds Elicits Deep AI Searcher
|
17 |
-
|
18 |
-
MindSearch is an open-source AI Search Engine Framework with Perplexity.ai Pro performance. You can simply deploy it with your own perplexity.ai style search engine with either close-source LLMs (GPT, Claude) or open-source LLMs (InternLM2.5-7b-chat). It owns following features:
|
19 |
-
|
20 |
-
- 🤔 **Ask everything you want to know**: MindSearch is designed to solve any question in your life and use web knowledge.
|
21 |
-
- 📚 **In-depth Knowledge Discovery**: MindSearch browses hundreds of web pages to answer your question, providing deeper and wider knowledge base answer.
|
22 |
-
- 🔍 **Detailed Solution Path**: MindSearch exposes all details, allowing users to check everything they want. This greatly improves the credibility of its final response as well as usability.
|
23 |
-
- 💻 **Optimized UI Experimence**: Providing all kinds of interfaces for users, including React, Gradio, Streamlit and Terminal. Choose any type based on your need.
|
24 |
-
- 🧠 **Dynamic Graph Construction Process**: MindSearch decomposes the user query into atomic sub-questions as nodes in the graph and progressively extends the graph based on the search result from WebSearcher.
|
25 |
-
|
26 |
-
<div align="center">
|
27 |
-
|
28 |
-
<img src="assets/teaser.gif">
|
29 |
-
|
30 |
-
</div>
|
31 |
-
|
32 |
-
## ⚡️ MindSearch vs other AI Search Engines
|
33 |
-
|
34 |
-
Comparison on human preference based on depth, breadth, factuality of the response generated by ChatGPT-Web, Perplexity.ai (Pro), and MindSearch. Results are obtained on 100 human-crafted real-world questions and evaluated by 5 human experts\*.
|
35 |
-
|
36 |
-
<div align="center">
|
37 |
-
<img src="assets/mindsearch_openset.png" width="90%">
|
38 |
-
</div>
|
39 |
-
* All experiments are done before July.7 2024.
|
40 |
-
|
41 |
-
## ⚽️ Build Your Own MindSearch
|
42 |
-
|
43 |
-
### Step1: Dependencies Installation
|
44 |
-
|
45 |
-
```bash
|
46 |
-
pip install -r requirements.txt
|
47 |
-
```
|
48 |
-
|
49 |
-
### Step2: Setup MindSearch API
|
50 |
-
|
51 |
-
Setup FastAPI Server.
|
52 |
-
|
53 |
-
```bash
|
54 |
-
python -m mindsearch.app --lang en --model_format internlm_server
|
55 |
-
```
|
56 |
-
|
57 |
-
- `--lang`: language of the model, `en` for English and `zh` for Chinese.
|
58 |
-
- `--model_format`: format of the model.
|
59 |
-
- `internlm_server` for InternLM2.5-7b-chat with local server. (InternLM2.5-7b-chat has been better optimized for Chinese.)
|
60 |
-
- `gpt4` for GPT4.
|
61 |
-
if you want to use other models, please modify [models](./mindsearch/agent/models.py)
|
62 |
-
|
63 |
-
### Step3: Setup MindSearch Frontend
|
64 |
-
|
65 |
-
Providing following frontend interfaces,
|
66 |
-
|
67 |
-
- React
|
68 |
-
|
69 |
-
```bash
|
70 |
-
# Install Node.js and npm
|
71 |
-
# for Ubuntu
|
72 |
-
sudo apt install nodejs npm
|
73 |
-
|
74 |
-
# for windows
|
75 |
-
# download from https://nodejs.org/zh-cn/download/prebuilt-installer
|
76 |
-
|
77 |
-
# Install dependencies
|
78 |
-
|
79 |
-
cd frontend/React
|
80 |
-
npm install
|
81 |
-
npm start
|
82 |
-
```
|
83 |
-
|
84 |
-
Details can be found in [React](./frontend/React/README.md)
|
85 |
-
|
86 |
-
- Gradio
|
87 |
-
|
88 |
-
```bash
|
89 |
-
python frontend/mindsearch_gradio.py
|
90 |
-
```
|
91 |
-
|
92 |
-
- Streamlit
|
93 |
-
|
94 |
-
```bash
|
95 |
-
streamlit run frontend/mindsearch_streamlit.py
|
96 |
-
```
|
97 |
-
|
98 |
-
## 🐞 Debug Locally
|
99 |
-
|
100 |
-
```bash
|
101 |
-
python -m mindsearch.terminal
|
102 |
-
```
|
103 |
-
|
104 |
-
## 📝 License
|
105 |
-
|
106 |
-
This project is released under the [Apache 2.0 license](LICENSE).
|
107 |
-
|
108 |
-
## Citation
|
109 |
-
|
110 |
-
If you find this project useful in your research, please consider cite:
|
111 |
-
|
112 |
-
```
|
113 |
-
@misc{chen2024mindsearchmimickinghumanminds,
|
114 |
-
title={MindSearch: Mimicking Human Minds Elicits Deep AI Searcher},
|
115 |
-
author={Zehui Chen and Kuikun Liu and Qiuchen Wang and Jiangning Liu and Wenwei Zhang and Kai Chen and Feng Zhao},
|
116 |
-
year={2024},
|
117 |
-
eprint={2407.20183},
|
118 |
-
archivePrefix={arXiv},
|
119 |
-
primaryClass={cs.CL},
|
120 |
-
url={https://arxiv.org/abs/2407.20183},
|
121 |
-
}
|
122 |
-
```
|
|
|
1 |
+
title: mindsearch
|
2 |
+
emoji: 😊
|
3 |
+
colorFrom: yellow
|
4 |
+
colorTo: yellow
|
5 |
+
sdk: streamlit
|
6 |
+
sdk_version: "1.34.0"
|
7 |
+
app_file: frontend/mindsearch_streamlit.py
|
8 |
+
pinned: false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|