ResearchGPT is an AI-powered research assistant that helps users find information from multiple sources using natural language queries. It combines LangChain's agent capabilities with Groq LLMs to provide comprehensive answers by searching across:
- 📚 Wikipedia for general knowledge
- 📑 arXiv for scientific papers and research
- 🌐 DuckDuckGo for web search results
- Python 3.8+
- Groq API key (get one at console.groq.com)
-
Clone this repository:
git clone https://github.com/yourusername/researchgpt.git cd researchgpt -
Install dependencies:
pip install -r requirements.txt
-
Run the application:
streamlit run app.py
-
Open your browser and navigate to
http://localhost:8501
You can set your Groq API key as an environment variable to avoid entering it each time:
export GROQ_API_KEY=your_api_key_here- Enter your Groq API key in the sidebar (if not set as an environment variable)
- Select your preferred LLM model
- Adjust tool settings if needed
- Type your research question in the chat input
- Explore the AI's response, which will show its reasoning process and the tools it used
- "What are the latest developments in quantum computing?"
- "Explain the impact of climate change on coral reefs"
- "How do mRNA vaccines work?"
- "What research exists on sleep and memory consolidation?"
- "Summarize recent papers on transformer neural networks"
This application uses:
- Streamlit: For the web interface and UI components
- LangChain: For agent orchestration and tool integration
- Groq: For fast LLM inference
- Wikipedia API: For encyclopedia knowledge
- arXiv API: For scientific papers
- DuckDuckGo Search: For web results
The agent uses a zero-shot reasoning approach to determine which tool to use based on the query content.