sqb dbt clone populates a target with the already-built relations from your production baseline, without running dbt. It resolves the production-shaped relations the same way reuse does (by compiling your dbt project at the configured reuse_from git ref), then clones or copies each one into the target schema.
reuse_from.git_ref from your configuration; the destination is the active target. Where the warehouse supports zero-copy cloning, relations are cloned (CLONED); otherwise they are copied (COPIED). Views are recreated to point at the cloned relations (RECREATED_VIEWS).
This differs from reuse from production, which runs as a pre-phase inside a build and only brings over the unchanged relations a build needs. sqb dbt clone is a standalone operation that populates the whole target (or a selected subset) up front.
Options
| Flag | Effect |
|---|---|
--select / -s | Limit to matching models (dbt selection syntax). |
--exclude | Exclude matching models. |
--hard-copy | Force a full copy instead of a zero-copy clone, even where the warehouse supports cloning. |
--no-sql-validation | Skip SQL validation during the clone. |
On this topic
- Reuse from production - reuse unchanged relations as a build pre-phase.
- Diff - compare a build against the production baseline.

