This project is designed to scrape a website or YouTube video by URL link and provide a concise summary to the user. It utilizes the power of OpenAI and GROQ API for language processing and LangChain for text and document handling.
- Content Summarization: Provides a summary of the content from a given URL.
- YouTube and Web Support: Handles both YouTube video URLs and regular web page URLs.
- Customizable Temperature: Allows setting the temperature for the language model to control the creativity of the output.
- Next.js: Backend framework.
- OpenAI: For generating embeddings and language processing.
- GROQ API: For language model interaction.
- LangChain: For handling text splitting, document combination, and retrieval.
- Puppeteer: For web scraping.
- YouTube Data API: For YouTube video data retrieval.
- React: Frontend framework.
- Axios: For making HTTP requests.
- Lucide React: For icons.
- Tailwind CSS: For styling.
- Node.js
- npm or yarn
- OpenAI API Key
- (Optional) GROQ API Key
- (Optional) Local LLM URL
Contributions are welcome! Please submit a pull request or open an issue to discuss any changes.
This project is licensed under the MIT License. See the LICENSE file for details.
- LangChain
- OpenAI
- Next.js
- Puppeteer
- React
- Axios
- Lucide React