Create a Multimodal AI Chat Assistant with GPT-4o for Text, Images, and PDFs

This workflow leverages the capabilities of OpenAI's GPT-4o model to create a chat assistant that can process and respond to text, images, and PDFs. It begins with a chat trigger node that initiates a session using n8n's hosted chat UI. Users can send text or upload files, which are then analyzed by the AI model. For images, the workflow converts them to base64 format for analysis. The AI generates natural language descriptions and maintains conversation context with a memory buffer, ensuring coherent follow-up interactions. This setup is modular, allowing for integration with various input methods and extensions like saving results to Notion or sending replies via Slack.