Integrating with Azure OpenAI in Orkes Conductor
To effectively utilize AI and LLM tasks in Orkes Conductor, it's essential to integrate your Conductor cluster with the necessary AI and LLM models.
Azure OpenAI offers a range of models that can be incorporated into the Orkes Conductor console. The choice of model depends on your unique use case, the functionalities you require, and the specific natural language processing tasks you intend to tackle.
This guide will provide the steps for integrating the Azure OpenAI provider with Orkes Conductor.
Steps to integrate with Azure OpenAI
Before beginning to integrate with Azure OpenAI, you need to generate the API key & get the endpoints from the Azure portal.
Get the Azure OpenAI API keys & Endpoint
- Navigate to the Azure portal.
- Create an account or sign up with your existing account.
- Under the Azure services section, choose Azure OpenAI.
- Go to your resource and navigate to Resource Management > Keys and Endpoint.
- Copy the API keys. You can either use KEY 1 or KEY 2 as the API Key.
- Copy the endpoint provided under the OpenAI section. An example endpoint is https://docs-test-001.openai.azure.com/.
Integrating with Azure OpenAI as a model provider
Now, you have the API keys & endpoint. Let’s integrate this with Orkes Conductor.
- Navigate to Integrations from the left menu on your Orkes Conductor console.
- Click +New integration button from the top-right of your window.
- Under the AI / LLM section, choose Azure + OpenAI.
- Click +Add and provide the following parameters:
|Integration name||Provide a name for the integration.|
|API Key||Provide the API key copied previously.|
|Endpoint||Provide the endpoint copied previously.|
|Description||Provide a description of your integration.|
- You can toggle-on the Active button to activate the integration instantly.
- Click Save.
Adding Azure OpenAI models to the integration
Now, you have integrated your Conductor console with the Azure OpenAI provider. The next step is integrating with the specific models.
Azure OpenAI has different models, such as text-davinci-003, text-embedding-ada-002, gpt-35-turbo and more. Each model is to be used for different use cases, such as text completion, generating embeddings, etc.
Depending on your use case, you must configure the required model within your Azure OpenAI configuration.
To add a new model to the Azure OpenAI integration:
- Navigate to the integrations page and click the '+' button next to the integration created.
- Click +New model.
- Provide the model name and an optional description for the model. You can get the complete list of models in Azure OpenAI here.
- Click the Enable checkbox to enable the model immediately.
- Click Save.
This ensures the integration model is saved for future use in LLM tasks within Orkes Conductor.
RBAC - Governance on who can use Integrations
Now, the integration with the required models is ready. Next, we should determine the access control to these models.
The permission can be granted to applications/groups within the Orkes Conductor console.
To provide explicit permission to Groups:
- Navigate to Access Control > Groups from the left menu on your Orkes Conductor console.
- Create a new group or choose an existing group.
- Under the Permissions section, click +Add Permission.
- Under the Integrations tab, select the required integrations with the required permissions.
- Click Add Permissions. This ensures that all the group members can access these integration models in their workflows.
Similarly, you can also provide permissions to applications.