Bridges Apache Solr search indexes with vector embeddings for hybrid keyword and semantic document retrieval, enabling contextual searches against structured data repositories without direct database access
Integrates with Florence-2 to enable advanced image analysis and manipulation tasks like visual question answering, image captioning, and content-based image retrieval.
Processes images and PDFs through Mistral AI's OCR API to extract text from visual documents, supporting both local files and URLs with Docker containerization for easy deployment.