OpenResearcher/OpenResearcher-30B-A3B-GGUF

Adopted by NVIDIA's Nemotron family of models!

🤗 HuggingFace ｜ Blog ｜ Slack | WeChat

Overview

OpenResearcher is a fully open agentic large language model (30B-A3B) designed for long-horizon deep research scenarios. It achieves an impressive 54.8% accuracy on BrowseComp-Plus, surpassing performance of GPT-4.1, Claude-Opus-4, Gemini-2.5-Pro, DeepSeek-R1 and Tongyi-DeepResearch. It also demonstrates leading performance across a range of deep research benchmarks, including BrowseComp, GAIA, WebWalkerQA, and xbench-DeepSearch. We fully open-source the training and evaluation recipe—including data, model, training methodology, and evaluation framework for everyone to progress deep research.

OpenResearcher-30B-A3B-GGUF

Note: For the best performance, we recommend using OpenResearcher-30B-A3B.

To support efficient deployment, we release several quantized versions of OpenResearcher-30B-A3B, including Q4_K_M, Q5_0, Q5_K_M, Q6_K, and Q8_0.

Quantization	File Size	BPW	PPL	+/-	Tokens/sec
BF16	58.84 GiB	16.00	8.4522	0.06489	4,117.90
Q8_0	31.27 GiB	8.51	8.4654	0.06499	7,490.81
Q6_K	31.20 GiB	8.49	8.4784	0.06510	7,389.76
Q5_0	20.37 GiB	5.54	8.5462	0.06558	7,534.66
Q4_K_M	22.82 GiB	6.21	8.5970	0.06610	7,046.96
Q5_K_M	24.24 GiB	6.60	8.6074	0.06625	6,661.48

Citation

@article{li2026openresearcher,
  title={{OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis}},
  author={Li, Zhuofeng and Jiang, Dongfu and Ma, Xueguang and Zhang, Haoxiang and Nie, Ping and Zhang, Yuyu and Zou, Kai and Xie, Jianwen and Zhang, Yu and Chen, Wenhu},
  journal={arXiv preprint arXiv:2603.20278},
  year={2026}
}