---
title: Deploy an LLM from the playground
description: LLM blueprints and all their associated settings are registered in the Registry and can be deployed and monitored with the Console.
section_name: Generative AI
maturity: public-preview
platform: cloud-only

---

# Deploy an LLM from the playground {: #deploy-an-llm-from-the-playground }

From an LLM [playground](playground) in a [Use Case](wb-usecase/index), add an LLM blueprint to the [Registry](nxt-registry/index) to prepare the LLM for use in production. After creating a draft LLM blueprint, setting the blueprint configuration (including the base LLM and, optionally, a system prompt and vector database), and testing and tuning the responses, the blueprint is ready for registration and deployment. Save the draft as a blueprint and add it to the Registry's model workshop as a custom model with the text generation target type:

1. In a **Use Case**, on the **Playgrounds** tab, click the playground containing the LLM you want to save as a blueprint.

2. Click the blueprint draft to save, click the actions menu (:material-dots-vertical:{.lg }), and then click **Save as LLM blueprint**. You can also click **Save as LLM blueprint** in **Draft actions** or in the :fontawesome-solid-gear:{.lg } **Configuration** sidebar.

    ![](images/playground-deploy-1.png)

    The blueprint draft is now an LLM blueprint. Once saved, no further changes can be made to the blueprint.

3. Click the actions menu (:material-dots-vertical:{.lg }), and then click **Register LLM blueprint**. You can also click **LLM blueprint actions > Register LLM blueprint**.

    ![](images/playground-deploy-2.png)

4. In the lower-left corner of the LLM playground, notifications appear as the LLM is queued and registered. When the registration notification appears, click **Go to the LLM blueprint in Registry**:

    ![](images/playground-deploy-notification.png)

    The LLM blueprint opens in the Registry's model workshop as a custom model with the **Text Generation** target type:

    ![](images/playground-deploy-3.png)

5. On the **Assemble** tab, in the **Runtime Parameters** section, configure the key-value pairs required by the LLM, including the LLM service's credentials and other details. To add these values, click the edit icon (:material-pencil:{.lg}) next to the available runtime parameters.

    To configure **Credential** type **Runtime Parameters**, first, add the credentials required for the LLM you're deploying to the [**Credentials Management** page of the DataRobot platform](stored-creds#credentials-management):

    === "Microsoft-hosted LLMs" 

        **Microsoft-hosted LLMs**: Azure OpenAI GPT-3.5 Turbo, Azure OpenAI GPT-3.5 Turbo 16k, Azure OpenAI GPT-4, and Azure OpenAI GPT-4 32k 

        **Credential type**: API Token (_not_ Azure)

        ![](images/playground-deploy-credential-token.png)

        The required **Runtime Parameters** are:

        ![](images/playground-deploy-4.png)

        Key                      | Description
        -------------------------|------------
        OPENAI_API_KEY           | Select the **API Token** credential, created on the [**Credentials Management** page](stored-creds#credentials-management), for the Azure OpenAI LLM API endpoint.
        OPENAI_API_BASE          | Enter the URL for the Azure OpenAI LLM API endpoint.
        OPENAI_API_DEPLOYMENT_ID | Enter the name of the Azure OpenAI deployment of the LLM, chosen when deploying the LLM to your Azure environment. For more information, see the Azure OpenAI documentation on how to [Deploy a model](https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/create-resource?pivots=web-portal#deploy-a-model){ target=_blank }. The default deployment name suggested by DataRobot matches the ID of the LLM in Azure OpenAI (for example, gpt-35-turbo). Modify this parameter if your Azure OpenAI deployment is named differently.
        OPENAI_API_VERSION       | Enter the Azure OpenAI API version to use for this operation, following the YYYY-MM-DD or YYYY-MM-DD-preview format (for example, 2023-05-15). For more information on the supported versions, see the [Azure OpenAI API reference documentation](https://learn.microsoft.com/en-us/azure/ai-services/openai/reference){ target=_blank }.
        PROMPT_COLUMN_NAME       | Enter the prompt column name from the input .csv file. The default column name is promptText.	

    === "Amazon-hosted LLMs" 

        **Amazon-hosted LLM**: Amazon Titan

        **Credential type**: AWS

        ![](images/playground-deploy-credential-amazon.png)

        The required **Runtime Parameters** are:

        ![](images/playground-deploy-parameters-amazon.png)

        Key                 | Description
        --------------------|------------
        AWS_ACCOUNT         | Select an **AWS** credential, created on the [**Credentials Management** page](stored-creds#credentials-management), for the AWS account.
        AWS_REGION          | Enter the AWS region of the AWS account. The default is us-west-1.
        PROMPT_COLUMN_NAME  | Enter the prompt column name from the input .csv file. The default column name is promptText.

    === "Google-hosted LLMs" 

        **Google-hosted LLM**: Google Bison

        **Credential type**: Google Cloud Service Account

        ![](images/playground-deploy-credential-google.png)

        The required **Runtime Parameters** are:

        ![](images/playground-deploy-parameters-google.png)

        Key                    | Description
        -----------------------|------------
        GOOGLE_SERVICE_ACCOUNT | Select a **Google Cloud Service Account** credential created on the [**Credentials Management** page](stored-creds#credentials-management).
        GOOGLE_REGION          | Enter the GCP region of the Google service account. The default is us-west-1.
        PROMPT_COLUMN_NAME     | Enter the prompt column name from the input .csv file. The default column name is promptText.

6. In the [**Settings** section](nxt-create-custom-model#configure-custom-model-resource-settings), ensure **Network access** is set to **Public**.

7. After you complete the custom model assembly, you can [test the model](nxt-test-custom-model) or [create new versions](nxt-custom-model-versions). DataRobot recommends testing custom LLMs before deployment.

8. Next, click **Register model**, [provide the registered model or version details](nxt-register-cus-models#register-a-model-from-the-model-workshop), and then click **Register model** again to add the custom LLM to the Registry.

    ![](images/playground-deploy-5.png)

    The registered model version opens in the **Model directory**.

9. From the **Model directory**, in the upper-right corner of the registered model version panel, click **Deploy** and [configure the deployment settings](nxt-deploy-models#configure-deployment-settings).

    ![](images/playground-deploy-6.png)

    For more information on the deployment functionality available for generative models, see [Monitoring support for generative models](generative-model-monitoring).

For more information on this process, see the [Playground deployment considerations](genai-consider#playground-deployment-considerations).