Configure BYOK for dbt Wizard Beta
The dbt Wizard CLI runs in bring-your-own-key (BYOK) mode, which means you supply credentials from a supported provider.
We'd love to hear how dbt Wizard is working for you. Share your feedback by either running the /feedback slash command in your interactive terminal session or by going to the #dbt-wizard channel in the dbt Community Slack.
Thanks so much for your help in improving dbt Wizard and dbt data development!
BYOK applies to the CLI only. The dbt platform uses a separate account-level integration.
BYOK works as follows:
- dbt Wizard calls your chosen provider's API directly using your key.
- Usage costs appear on your provider account, so token costs are billed directly by whichever provider you choose.
Supported AI providers
dbt Wizard
dbt Wizard supports different AI providers depending on where you use it.
- Configure dbt platform integrations in account settings. BYOK is available for Enterprise and Enterprise+ plans.
- Configure BYOK for the CLI. For OpenAI subscription support in the CLI, run
wizard providers configure openai_subscriptionand follow the prompts.
Configure with wizard providers
Use the providers subcommand to enable a provider and store credentials:
wizard providers list
wizard providers configure openai_subscription # or openai, anthropic, bedrock, azure, snowflake
wizard providers enable openai_subscription
You can also configure providers interactively from an active CLI session with the /providers slash command.
For OpenAI subscription support, run wizard providers configure openai_subscription and follow the prompts. Use wizard providers configure openai if you want to use an OpenAI API key instead.
To store an API key without echoing it in your shell history:
printf '%s' 'sk-...' | wizard providers set-key openai
Credentials are stored in ~/.dbt/wizard/provider-auth.json. Provider settings are stored in ~/.dbt/wizard/providers.json.
Set your API key
- Interactive session
- Environment variable
The first time you start dbt Wizard in a project, onboarding prompts you to choose a provider:
- At the Configure a Provider prompt, select your provider (OpenAI subscription, OpenAI API key, Anthropic, Amazon Bedrock, Azure, or Snowflake).
- Paste your API key or cloud credential details when prompted, then press Enter.
- Choose an AI model to finish.
To add or switch providers later from an active session, type /providers in the TUI. dbt Wizard stores credentials in ~/.dbt/wizard/provider-auth.json.
Set the key as an environment variable — the simplest option for headless runs (like wizard exec) and to reuse your key across sessions:
- OpenAI
- Anthropic
export OPENAI_API_KEY="sk-..."
export ANTHROPIC_API_KEY="sk-ant-..."
Add this line to your shell profile (.zshrc, .bashrc, or equivalent) to make it permanent across terminal sessions.
To persist provider credentials, use wizard providers configure PROVIDER_NAME, wizard providers set-key PROVIDER_NAME, or the provider's environment variable.
Choose an AI model
You can change the AI model in the following ways:
Make sure to restart the dbt Wizard CLI to apply the change.
AWS Bedrock
AWS Bedrock is supported in the CLI only. Wizard uses your local AWS credential chain (environment variables, shared credentials file, or IAM role).
-
Enable the Bedrock provider:
wizard providers enable bedrock -
Set the AWS Region (and optional profile):
wizard providers bedrock set-region us-east-1
wizard providers bedrock set-profile default # optional -
Or run the interactive setup:
wizard providers configure bedrock -
Choose a Bedrock AI model with
/modelin the TUI, or set a default inconfig.toml. Runwizard debug modelsto list available model IDs.
Ensure your AWS account has access to the Bedrock models you plan to use.
Snowflake Cortex (preview)
Snowflake Cortex BYOK support in the CLI is in preview. Availability and setup steps may change.
-
Enable the Snowflake provider when supported in your CLI version:
wizard providers enable snowflake -
Run the interactive setup:
wizard providers configure snowflake -
Ensure your Snowflake account has the privileges required for Cortex LLM functions. Refer to the Snowflake Cortex documentation.
Run wizard providers list to confirm whether Cortex is available in your installed version.
Related docs
Was this page helpful?
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.