Introducing FILE type: a native column type for multimodal data
Summary
The new FILE column type, now in beta, lets you store unstructured data like documents, images, audio, and video natively in your tables. It enables unified governance using standard table access controls and security policies, with community support underway across Delta Lake, Parquet, Iceberg, and Spark for broad portability.
Summary generated by brickster.ai. For the full article, follow the source link above.
Topics
More from Databricks Blog
Your data, your storage, your rules: a 2026 guide to storing Unity Catalog managed tables
You can redirect where Unity Catalog managed tables land by using SET MANAGED LOCATION and moving existing data through external table conversion. Defining managed storage paths at the metastore, catalog, or schema level keeps you in control of your cloud storage to support compliance and accurate cost attribution.
How Databricks rolls out frontier models to 12,000 employees on Day 1
Databricks delivers Day 1 access to frontier AI models for all 12,000 employees to ensure cutting-edge capabilities are immediately available internally. See how the company prioritizes and operationalizes large-scale model rollouts across its entire workforce.
How Databricks rolls out frontier models to 12,000 employees on Day 1
Databricks makes workforce access to frontier AI capabilities a top priority by rolling out newly released models to 12,000 employees on Day 1. This organization-wide rollout demonstrates how the platform delivers immediate, day-one access to cutting-edge AI capabilities at scale.
Lakebase Search: State-of-the-art full text and vector search for Postgres
Lakebase Postgres now includes a built-in search engine, generally available on AWS and Azure, that enables native semantic, keyword, and hybrid search directly alongside operational data without separate ETL pipelines. Featuring a serverless architecture that scales to zero and decouples storage from compute, it delivers twice the throughput at a quarter of the cost of cloud Postgres with pgvector on 100-million-vector benchmarks.
Manufacturing data and AI: Connecting the product value chain
Databricks connects fragmented manufacturing systems across functions and plants by unifying data or querying it in place to answer cross-stage value chain questions. Governed semantics, natural-language analytics, and agentic applications empower teams to move from finding data to taking action without becoming data engineers.
How to roll out Genie One: A step-by-step enterprise playbook
Scaling Genie One from an initial demo to trusted enterprise adoption requires starting with a single, well-governed data domain. This step-by-step playbook outlines a phased plan to expand the data-smart AI coworker from a pilot team to org-wide adoption without losing control of your data.