ClickHouse is now available on the dbt platform

Introduction

The integration of ClickHouse with the dbt platform marks a significant milestone in the evolution of data engineering tools. As of September 16, 2026, the ClickHouse adapter for dbt v2 is available in public beta, leveraging the Rust-based rewrite of dbt. This update brings performance improvements and a more streamlined integration process. Additionally, ClickHouse is now available on the dbt platform in private beta, supporting both open-source ClickHouse and ClickHouse Cloud.

Key Features and Benefits

The ClickHouse adapter for dbt v2 is built on the Rust engine, which offers notable performance gains compared to the previous Python-based version. This new adapter is included in the dbt binary, eliminating the need for separate installation. The integration with the dbt platform allows users to create ClickHouse connections, develop models in Studio, test them across environments, and run scheduled jobs with logs and retries.

The dbt platform now supports ClickHouse as a natively integrated data warehouse, making it the first partner-built v2 adapter available on the platform. This integration is particularly beneficial for teams that require real-time analytics and high concurrency, as ClickHouse is designed for sub-second queries on recently ingested data.

Integration and Usage

The ClickHouse adapter for dbt v2 is available in public beta and can be installed via pip. Users can request access to the private beta form to start using ClickHouse on the dbt platform. The integration process involves setting up a ClickHouse connection in the dbt platform, which uses the dbt v2 adapter and supports the same beta limitations as the local adapter.

For teams using ClickHouse Cloud, the dbt platform provides a direct connection, allowing users to develop and test models in Studio, run scheduled jobs, and browse models in Catalog. This integration simplifies the workflow for data teams, enabling them to leverage the power of ClickHouse within the dbt ecosystem.

Performance and Scalability

The dbt v2 adapter is a ground-up rewrite of dbt in Rust, offering significant performance improvements. dbt Labs reports parse times up to 30x faster than dbt v1, reducing compilation times from minutes to seconds. This improvement enhances the feedback loop in the editor, making it easier for developers to stay in flow.

ClickHouse's ability to handle real-time workloads is complemented by the dbt v2 adapter's performance gains. The combination of ClickHouse's sub-second query capabilities and dbt's efficient model execution makes it an attractive option for teams requiring real-time analytics and high concurrency.

Limitations and Considerations

While the dbt v2 adapter for ClickHouse is available in public beta, it is not yet ready for production workloads. It is recommended to use the beta in development or staging environments on ClickHouse Cloud or a single-node self-managed instance. The private beta on the dbt platform also has limitations, and users are encouraged to provide feedback during this phase to help shape the final product.

Additionally, some features are still missing in the v2 adapter, and users should review the v1 vs v2 parity table before testing an existing project. The dbt v2 adapter is expected to reach feature parity with dbt-clickhouse shortly, and users are encouraged to open issues in the repository if they encounter any discrepancies.

Conclusion

The integration of ClickHouse with the dbt platform represents a significant step forward in the data engineering landscape. The availability of the ClickHouse adapter for dbt v2 in public beta and the private beta on the dbt platform provide users with new opportunities to leverage the power of ClickHouse within the dbt ecosystem. As the dbt v2 adapter continues to evolve, it is expected to become a robust and reliable tool for data teams requiring real-time analytics and high concurrency.

Source: ClickHouse. Analysis based on publicly available information; no hands-on testing is claimed.