No module named 'dbt.adapters.snowflake'" install error

TL;DR

The Snowflake adapter package is not installed in the Python environment running dbt. Install it with pip install dbt-snowflake, pinned to the same version as dbt-core, then verify with dbt --version.

Error

"No module named 'dbt.adapters.snowflake'" install error

Steps

1. Check what is installed: pip show dbt-core and pip show dbt-snowflake. Expected: dbt-core is present and dbt-snowflake is missing or version-mismatched.
2. Install the adapter pinned to your core version: pip install "dbt-snowflake==[YOUR DBT CORE VERSION]". Expected: pip installs without dependency conflicts.
3. Verify the install: dbt --version. Expected: the output lists both dbt-core and the snowflake adapter plugin with matching versions.
4. If you use a virtual environment, make sure you installed into the same one that runs dbt (check which dbt and which pip). Expected: both point at the same environment.
5. Run dbt debug. Expected: the adapter loads and the connection check runs.

When to use

• dbt fails with No module named 'dbt.adapters.[ADAPTER]'.
• Fresh machine, new virtualenv, or new CI image.

When not to use

• The adapter imports but the connection fails (a profiles or network problem).
• You need a different adapter (install dbt-bigquery, dbt-postgres, etc. instead).

Tool compatibility

• dbt Core 1.x with the matching adapter major version. Always keep core and adapter versions in sync.

Variant phrasings

No module named 'dbt.adapters.bigquery' / 'dbt.adapters.postgres'

Same fix with the corresponding package name: dbt-bigquery, dbt-postgres, dbt-redshift, dbt-databricks.

Adapter version mismatch warnings

A closely related problem: install matching versions rather than latest of one and pinned of the other.

Why it happens

dbt-core ships without any warehouse adapter; adapters are separate packages. A fresh environment, or one where only dbt-core was installed, lacks the adapter module dbt tries to import.

Edge cases

• Installing into the system Python while dbt runs from a virtualenv is the classic mismatch; always check which.
• Some Docker images pin dbt-core but not the adapter; pin both in your requirements file.
• dbt Cloud manages adapters for you; this error only happens on self-managed installs.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_GrluVDtyB5yASCDzjWV-xQ