Overview Quickstart dbt Compatibility The Style Guide profiles.yml sources.yml Configuration Supported Databases

DUCKDB

In-process analytical database. DVT uses DuckDB internally for federation compute and as the extraction cache engine.

DVT SUPPORT

✓ FEDERATION SOURCE

read in federation models

✓ FEDERATION TARGET

materialize federation results

✓ DEFAULT TARGET

run full dbt projects against it

✓ SEED TARGET

dvt seed --target loads here

DVT supports DuckDBbecause both dbt and Sling do: dbt drives standard materializations when it's your default target, Sling moves data in and out during federation. Connectivity for both layers is checked with dvt debug --all.

PREREQUISITES

Driver: duckdb — installed automatically by:

dvt sync

CONFIGURATION FIELDS

FIELD	TYPE	REQUIRED	DEFAULT	DESCRIPTION
type	string	yes	—	Must be `duckdb`
path	string	no	:memory:	Database file path (or :memory: for in-memory)
threads	integer	no	4	Number of parallel threads

PROFILES.YML EXAMPLE

my_project:
  target: duckdb_dev
  outputs:
    duckdb_dev:
      type: duckdb
      path: ./analytics.duckdb

SOURCES.YML EXAMPLE

sources:
  - name: local_files
    meta:
      connection: duckdb_dev
    tables:
      - name: parquet_data

DBT ADAPTER FEATURES

When DuckDB is your default target, every standard dbt model runs through the official dbt adapter — all of its documented capabilities apply unchanged:

MATERIALIZATIONS

INCREMENTAL STRATEGIES

NOTABLE CONFIGS

▸external (Parquet/CSV) materializations
▸also DVT's federation compute engine

Official dbt DuckDB adapter docs ↗

FEDERATION INCREMENTAL STRATEGIES

Strategies available when this engine is the target of a f_incremental model:

✓ Append✓ Delete+Insert✓ Merge

KNOWN LIMITATIONS

⚠Single-user — file locks prevent concurrent access