Instructions to use OpenResearcher/OpenResearcher-30B-A3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenResearcher/OpenResearcher-30B-A3B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="OpenResearcher/OpenResearcher-30B-A3B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OpenResearcher/OpenResearcher-30B-A3B", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("OpenResearcher/OpenResearcher-30B-A3B", trust_remote_code=True, device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use OpenResearcher/OpenResearcher-30B-A3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenResearcher/OpenResearcher-30B-A3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenResearcher/OpenResearcher-30B-A3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/OpenResearcher/OpenResearcher-30B-A3B

SGLang

How to use OpenResearcher/OpenResearcher-30B-A3B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "OpenResearcher/OpenResearcher-30B-A3B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenResearcher/OpenResearcher-30B-A3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "OpenResearcher/OpenResearcher-30B-A3B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenResearcher/OpenResearcher-30B-A3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use OpenResearcher/OpenResearcher-30B-A3B with Docker Model Runner:
```
docker model run hf.co/OpenResearcher/OpenResearcher-30B-A3B
```

OpenResearcher-30B-A3B / README.md

ZhuofengLi

Update README.md

f056376 verified 30 days ago

preview code

Raw

History Blame Contribute Delete

8.02 kB

metadata

license: mit
datasets:
  - OpenResearcher/OpenResearcher-Dataset
base_model:
  - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16
library_name: transformers

Adopted by NVIDIA's Nemotron family of models!

🤗 HuggingFace ｜ Slack | WeChat

OpenResearcher-30B-A3B Overview

OpenResearcher-30B-A3B is an agentic large language model designed for long-horizon deep research fine-tuned from NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16 on 96K OpenResearcher dataset with 100+ turns. The dataset is derived by distilling GPT-OSS-120B with native browser tools. More info can be found on the dataset card at OpenResearcher dataset.

The model 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.

Deep Research Benchmark Results

Evaluate OpenResearcher-30B-A3B

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.

Quick Start

We provide a quick-start in GitHub that demonstrates how to use OpenResearcher-30B-A3B for deep research.

Core Contributors

Advisors

_{Wenhu Chen}

_{Yu Zhang}

Acknowledgements

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}
}