Worked on expanding multimodal data ingestion capabilities within the opea-project/GenAIComps and opea-project/GenAIExamples repositories, enabling the systems to handle image and audio files alongside existing video support. Developed and integrated new API endpoints, extended backend services, and updated documentation to support these modalities, using Python, Docker, and Shell scripting. Synchronized configuration and interface changes across both repositories to ensure consistency and facilitate future enhancements. The work established a robust foundation for multimodal question answering by allowing ingestion, processing, and querying of diverse data types, with no major defects reported during the period, and positioned the project for future expansion.