During April 2026, contributed to the langgenius/dify-official-plugins repository by developing vision-enabled text embedding and multimodal chat embedding features. This work enabled the processing of both text and image inputs, introducing a new chat embedding function that supports multimodal content for more versatile chat interactions. Leveraged Python and machine learning techniques, integrating vLLM and ensuring compatibility with OpenAI forms to optimize vision support and improve embedding accuracy and throughput. The release included a version update to reflect these enhancements, focusing on delivering richer multimodal capabilities and smoother downstream integrations. No bug fixes were recorded during this period, emphasizing feature delivery.