Real Postgres
Wire-compatible with standard Postgres. Same drivers, extensions, and ORMs you already use.
Lakebase. Managed Postgres, built for modern agentic workloads.
The serverless database with branching, autoscaling, and your Lakehouse data.
[Benefits]
Built for shipping, not provisioning. No infrastructure to stand up, no reverse-ETL to wire, no credentials to rotate. Just the database and your app.
[Benefits]
Lakehouse-native
Two-way Postgres-Lakehouse sync. Curated data for Lakebase app reads.
Auth, handled
Postgres roles linked to Databricks identities with short-lived secure tokens in apps.
[Features]
How Lakebase powers the apps you ship day to day
01Branching
Branch your database instantly. [A full-fidelity copy, regardless of size, fully isolated from its parent.]
Branches are copy-on-write — they share storage with their parent and only consume new space for the bytes you change.
- Per-PR preview environments
- Schema migration sandboxes
- Branch from any point in time
02Autoscaling
Right-size your compute automatically. [Demand-driven, scaled to zero on idle, milliseconds to wake.]
Lakebase tracks load in real time and adjusts capacity within the range you set, with no compute cost while idle.
- Non-disruptive scaling within range
- Independent autoscaling per replica
- Configurable idle timeout per branch
03Lakehouse sync
Connect your app to your Lakehouse. [Inbound and outbound, fully managed, governed by Unity Catalog.]
Both directions are managed by Databricks, no external pipelines, no jobs you have to operate, no glue code to maintain.
- Snapshot, triggered, or continuous sync
- Schema-level config for outbound replication
- Federated queries across both sides
[Use cases]
How teams use Lakebase in production. [Lakebase powers the data layer between apps, agents, and your Lakehouse.]
Shipping Full-Stack Apps
Use Lakebase as your app database for users, sessions, and logic, no external Postgres needed.
Powering Stateful AI Agents
Store conversations, tool outputs, and state so agents persist across sessions.
Serving Product Data
Bring Lakehouse data into Postgres for low-latency reads across APIs, ORMs, and apps.
Closing the Data Loop
Capture app writes for analytics with no custom pipelines needed.
Testing Database Changes
Validate schema changes and new features in isolated environments before they reach production.
Scaling Read-Heavy Apps
Handle high query volume by distributing reads without changing your application architecture.
Isolating Customers per Tenant
Run separate database environments per tenant for independent scaling.
Recovering and Debugging
Restore past data states to investigate issues and understand how your system evolved over time.
[Testimonials]
Lakebase powers applications. See how teams use it to bring data directly into user experiences.
"At Tibber, empowering customers to take control of their energy consumption requires a flexible data infrastructure. Lakebase's integration with Databricks makes it easy to serve analytical and transactional data, helping us deliver real-time insights to our customers."
Niklas Nordansjo, Data Platform Lead
"Lakebase lets an agentic team quickly self-serve the data they need for their models, whether it's historical claims or real-time transactions, and that's really powerful."
Dragon Sky, Chief Architect
"Lakebase gives us a durable, low-latency store for application state, so our data apps load quickly, refresh seamlessly and even support shared page links between users."
Bobby Muldoon, VP of Engineering
Databricks Developer Hub