Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
datasets:
|
| 4 |
+
- OpenResearcher/OpenResearcher-Dataset
|
| 5 |
+
base_model:
|
| 6 |
+
- nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16
|
| 7 |
+
---
|
| 8 |
+
<div style="display: flex; align-items: center; justify-content: center; gap: 8px;">
|
| 9 |
+
<img src="imgs/or-logo1.png" style="height: 84px; width: auto;">
|
| 10 |
+
<img src="imgs/openresearcher-title.svg" style="height: 84px; width: auto;">
|
| 11 |
+
</div>
|
| 12 |
+
|
| 13 |
+
<div align="center">
|
| 14 |
+
<a href="https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Dataset"><img src="https://img.shields.io/badge/Dataset-FFB7B2?style=for-the-badge&logo=huggingface&logoColor=ffffff" alt="Dataset"></a>
|
| 15 |
+
<a href="https://huggingface.co/OpenResearcher/Nemotron-3-Nano-30B-A3B"><img src="https://img.shields.io/badge/Model-FFD966?style=for-the-badge&logo=huggingface&logoColor=ffffff" alt="Model"></a>
|
| 16 |
+
<a href="https://boiled-honeycup-4c7.notion.site/OpenResearcher-A-Fully-Open-Pipeline-for-Long-Horizon-Deep-Research-Trajectory-Synthesis-2f7e290627b5800cb3a0cd7e8d6ec0ea?source=copy_link"><img src="https://img.shields.io/badge/Blog-4285F4?style=for-the-badge&logo=google-chrome&logoColor=white" alt="Blog"></a>
|
| 17 |
+
<a href="https://wandb.ai/dongfu/nano-v3-sft-search"><img src="https://img.shields.io/badge/WandB%20Logs-48B5A3?style=for-the-badge&logo=weightsandbiases&logoColor=white" alt="WandB Logs"></a>
|
| 18 |
+
<a href="https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Eval-Logs/tree/main"><img src="https://img.shields.io/badge/Eval%20Logs-755BB4?style=for-the-badge&logo=google-sheets&logoColor=white" alt="Eval Logs"></a>
|
| 19 |
+
</div>
|
| 20 |
+
|
| 21 |
+
</div>
|
| 22 |
+
<p align="center">
|
| 23 |
+
<img src="imgs/github.svg" width="15px" style="display:inline;"> <a href="https://github.com/TIGER-AI-Lab/OpenResearcher" target="_blank">Github</a> | 🤗 <a href="https://huggingface.co/collections/TIGER-Lab/openresearcher" target="_blank">HuggingFace</a> | <img src="imgs/notion.svg" width="15px" style="display:inline;"> <a href="https://boiled-honeycup-4c7.notion.site/OpenResearcher-A-Fully-Open-Pipeline-for-Long-Horizon-Deep-Research-Trajectory-Synthesis-2f7e290627b5800cb3a0cd7e8d6ec0ea?source=copy_link" target="_blank">Blog</a> | <img src="imgs/slack.png" width="14px" style="display:inline;"> <a href="https://join.slack.com/t/openresearcher/shared_invite/zt-3p0r32cky-PqtZkVjjWIAI14~XwcRMfQ" target="_blank">Slack</a> | <img src="imgs/wechat.svg" width="14px" style="display:inline;"> <a href="imgs/wechat_group.png" target="_blank">WeChat</a>
|
| 24 |
+
|
| 25 |
+
</p>
|
| 26 |
+
|
| 27 |
+
## OpenResearcher-30B-A3B Overview
|
| 28 |
+
OpenResearcher-30B-A3B is a language model fine-tuned from NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16 on 96K [OpenResearcher dataset](https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Dataset). The dataset is derived by distilling GPT-OSS-120B with [native browser tools](https://docs.vllm.ai/projects/recipes/en/latest/OpenAI/GPT-OSS.html#usage:~:text=Limitation%20section%20below.-,Tool%20Use,-%C2%B6). More info about the dataset can be found on the dataset card at [OpenResearcher dataset](https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Dataset.
|
| 29 |
+
|
| 30 |
+
The model achieves an impressive **54.8%** accuracy on [BrowseComp-Plus](https://huggingface.co/spaces/Tevatron/BrowseComp-Plus), surpassing performance of `GPT-4.1`, `Claude-Opus-4`, `Gemini-2.5-Pro`, `DeepSeek-R1` and `Tongyi-DeepResearch`.
|
| 31 |
+
<div align="center">
|
| 32 |
+
<img src="imgs/teaser.png" alt="OpenResearcher Teaser" width="100%" style="max-width: 850px; border-radius: 8px; box-shadow: 0 4px 10px rgba(0,0,0,0.1);">
|
| 33 |
+
</div>
|
| 34 |
+
|
| 35 |
+
<br>
|
| 36 |
+
## Deep Research Benchmark Results
|
| 37 |
+
|
| 38 |
+
<div align="center">
|
| 39 |
+
<img src="imgs/main_table.png" alt="Deep Research Benchmark Results" width="100%">
|
| 40 |
+
</div>
|
| 41 |
+
|
| 42 |
+
## Evaluate OpenResearcher-30B-A3B
|
| 43 |
+
We evaluate OpenResearcher-30B-A3B across a range of deep research benchmarks, including BrowseComp-Plus, BrowseComp, GAIA, xbench-DeepSearch. Please find more details in [GitHub](https://github.com/TIGER-AI-Lab/OpenResearcher)
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
## Quick Start
|
| 47 |
+
|
| 48 |
+
TODO
|
| 49 |
+
|
| 50 |
+
## Citation
|
| 51 |
+
```bibtex
|
| 52 |
+
@misc{li2025openresearcher,
|
| 53 |
+
title={OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis},
|
| 54 |
+
author={Zhuofeng Li and Dongfu Jiang and Xueguang Ma and Haoxiang Zhang and Yuyu Zhang and Kai Zou and Ping Nie and Jianwen Xie and Yu Zhang and Wenhu Chen},
|
| 55 |
+
year={2025},
|
| 56 |
+
howpublished={\url{https://www.notion.so/OpenResearcher-A-Fully-Open-Pipeline-for-Long-Horizon-Deep-Research-Trajectory-Synthesis-2f7e290627b5800cb3a0cd7e8d6ec0ea}},
|
| 57 |
+
note={Notion Blog}
|
| 58 |
+
}
|
| 59 |
+
```
|
| 60 |
+
|