GrokGazer is a cutting-edge, multimodal web intelligence explorer powered by the advanced AI capabilities of Grok 3, developed by xAI. Designed to empower students, coders, programmers, researchers, and innovators, GrokGazer tackles real-world challenges by transforming the way users interact with web data and documents. Leveraging Grok's state-of-the-art natural language processing and retrieval-augmented generation (RAG), this application extracts, analyzes, and synthesizes information with unprecedented efficiency, enabling users to unlock actionable insights from vast digital landscapes.
- For Students: Streamlines research by scraping and summarizing academic websites, helping with thesis work, study materials, and exam preparation.
- For Coders and Programmers: Accelerates code documentation and bug tracking by crawling repositories or forums, extracting keywords, and providing context-aware answers (helps in training their custom LLMs).
- For Researchers: Enhances data collection from diverse sources, offering structured outputs for analysis and hypothesis validation.
- For Innovators: Facilitates rapid prototyping by analyzing technical documents and web resources, fostering creativity with AI-driven insights.
By harnessing Grok 3's llama-3.3-70b-versatile model, GrokGazer delivers high-fidelity summarization, keyword extraction, and question-answering, optimized for scalability and precision. Integrated with PocketGroq's web crawling tools, it redefines how professionals and learners process information in an AI-augmented era.
- Web Scraping: Extract content from a single URL in Markdown, HTML, or structured data formats.
- Website Crawling: Navigate multiple pages with customizable depth and filters, outputting in Markdown, HTML, or structured data.
- Multimodal Analysis: Process uploaded PDFs, text files, and images, extracting text for AI-driven insights.
- AI-Powered Insights: Utilize Grok 3's AI to summarize content, extract keywords, and answer questions based on scraped or uploaded data.
- Download Options: Export results as JSON for offline use.
- User Interface: Streamlit-based UI with a sidebar for mode selection (Scrape, Crawl, Multimodal) and options, plus tabs for output viewing.
- Cache Management: Clear session state cache to optimize memory usage.
- About Section: Access detailed app information via a sidebar expander.
Future Enhancements: Audio mode for voice input and response playback is planned to further enhance accessibility and interaction.
- Python 3.11+
- Git (for cloning the repository)
- Internet connection (for API calls and web scraping)
- Microphone and speakers (for future audio mode)
- Ollama server (for local model management)
-
Clone the Repository:
git clone https://github.com/yourusername/GroqGazer.git cd GroqGazer -
Set Up Environment:
- Install Python dependencies:
`pip3 install --user -r requirements.txt` -
Configure API Keys:
- Create a .env file in the project root:
`GROQ_API_KEY=your_groq_api_key_here`- Obtain a Groq API key from https://console.groq.com.
-
Pull Ollama Models:
- Install Ollama: Follow instructions at https://ollama.com.
- Pull required models:
`ollama pull llama3 ollama pull nomic-embed-text` -
Run the Application:
- Start the Streamlit app:
`streamlit run groqgazer.py --server.enableCORS false --server.enableXsrfProtection false`- Open your browser at http://localhost:8501 to use GrokGazer.
- Scrape Mode: Enter a URL and select output formats to scrape content.
- Crawl Mode: Set depth and page limits to crawl a website.
- Multimodal Mode: Upload files (PDF, TXT, PNG, JPG) for analysis.
- Q&A: Ask questions based on extracted content.
- Cache: Clear cache via the sidebar button to manage memory.
- About: Expand the sidebar for app details.
Contributions are welcome! Please fork the repository, create a feature branch, and submit a pull request with your changes.
MIT License - Feel free to modify and distribute.
For issues or questions, open an issue on GitHub or contact the maintainers.
Built with ❤️ using Streamlit, PocketGroq, Groq, and xAI's Grok 3 technology.