### Docker setup for OmniThink

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Clone the repository, pull the Docker image, and run the container.

```bash
git clone https://github.com/zjunlp/OmniThink.git
docker pull zjunlp/omnithink:latest
docker run -it zjunlp/omnithink:latest
```

--------------------------------

### Install local search dependencies

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Install additional packages required for local search functionality.

```bash
# Install local search dependencies
pip install sentence-transformers faiss-cpu numpy

# Or use the updated requirements.txt
pip install -r requirements.txt
```

--------------------------------

### Conda environment setup for OmniThink

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Create a Conda environment with Python 3.11, clone the repository, and install requirements.

```bash
conda create -n OmniThink python=3.11
git clone https://github.com/zjunlp/OmniThink.git
cd OmniThink
# Install requirements
pip install -r requirements.txt
```

--------------------------------

### Local search setup with OfflineRAGFlow and LocalSearch

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Initialize the local RAG engine, ingest documents, create a DSPy-compatible search interface, and use it in a pipeline.

```python
from src.tools.rm import OfflineRAGFlow, LocalSearch

# Initialize the local RAG engine
rag_engine = OfflineRAGFlow(
    model_name="sentence-transformers/all-MiniLM-L6-v2",
    chunk_size=800,
    overlap=120,
    k=5
)

# Add documents to your local index
rag_engine.ingest(
    text="Your document content here...",
    meta={"title": "Document Title", "doc_id": "doc1"}
)

# Create DSPy-compatible search interface
local_search = LocalSearch(search=rag_engine, k=3)

# Use in your DSPy pipeline
results = local_search.forward("your search query")
```

--------------------------------

### Clone FActScore repository for evaluation

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Required setup before running any evaluation scripts. Clones the FActScore library into the eval directory.

```bash
cd eval
git clone https://github.com/shmsw25/FActScore.git
```

--------------------------------

### Export LM and search API keys

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Set environment variables for the LM API key and search key before running.

```bash
export LM_KEY=YOUR_API_KEY
export SEARCHKEY=YOUR_SEARCHKEY
```

--------------------------------

### Run OmniThink article generation with sh run.sh

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Executes the main OmniThink pipeline with a single command. Outputs the generated article, outline, and mindmap to the ./results/ directory.

```bash
sh run.sh
```

--------------------------------

### Run Knowledge Density evaluation

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Assesses knowledge density. Requires article path, API path, and number of threads.

```bash
python Knowledge_Density.py \
  --articlepath articlepath \
  --api_path api_path \
  --threads threads
```

--------------------------------

### Run Information Diversity evaluation

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Measures information diversity using a mapping file and a model path.

```bash
python Information_Diversity.py \
  --mappath mappath \
  --model_path model_path
```

--------------------------------

### Run Rubric Grading evaluation

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

Evaluates article quality using rubric grading. Requires paths to the article and the model.

```bash
python Rubric_Grading.py \
  --articlepath articlepath \
  --modelpath modelpath
```

--------------------------------

### Cite OmniThink with BibTeX entry

Source: https://github.com/zjunlp/omnithink/blob/main/README.md

BibTeX citation for OmniThink paper, useful for academic referencing.

```angular2
@misc{xi2025omnithinkexpandingknowledgeboundaries,
      title={OmniThink: Expanding Knowledge Boundaries in Machine Writing through Thinking}, 
      author={Zekun Xi and Wenbiao Yin and Jizhan Fang and Jialong Wu and Runnan Fang and Ningyu Zhang and Jiang Yong and Pengjun Xie and Fei Huang and Huajun Chen},
      year={2025},
      eprint={2501.09751},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2501.09751}, 
}
```