Introducing FILE type: a native column type for multimodal data
Databricks introduces a native FILE column type for storing multimodal data in tables
Databricks is adding a native FILE column type to its lakehouse platform, enabling structured tables to natively hold unstructured multimodal data like images, audio, and documents. This closes the gap between structured and unstructured data management, a key requirement for enterprise AI pipelines. It signals the data platform layer is evolving to meet multimodal AI workloads natively rather than relying on external blob stores.