Standalone pipelines

Standalone pipelines is a Databricks data engineering capability, introduced May 2026.

Declarative pipelines - materialized views and streaming tables - created from a notebook on serverless general compute, in SQL or Python.

• Renamed from "Pipelines for Databricks SQL," they let you build materialized views and streaming tables straight from a notebook on serverless general compute, in SQL or Python, with no dedicated pipeline cluster to stand up. 🔗
• They were previously called "Pipelines for Databricks SQL" - renamed once they could also run from a notebook on serverless general compute, not just a SQL warehouse. 🔗

⚠ Limitations: On serverless general compute: only the table owner can refresh; synchronous refresh only (no async); current channel only (no preview); a table refreshes only with the compute type it was created with; and cost attribution, vertical autoscaling on out-of-memory errors, schema-upgrade retries, and performance-mode selection aren't available. 🔗

Category

Data engineering

Introduced

May 2026🔗

Also known as

DBSQL pipelines, Pipelines on Databricks SQL

Verified

2026-07-23