Part 1: Creating an Azure OpenAI Instance
Navigate to the Azure Portal OpenAI creation page: https://portal.azure.com/#create/Microsoft.CognitiveServicesOpenAI
Create a new resource group and provide a name for your resource. Select your preferred region and choose the appropriate pricing tier for your needs.
In the network settings section, select "All networks including internet can access this." Proceed through the remaining steps and wait for the deployment to complete.
Part 2: Deploying and Configuring in Team-GPT
Access the Azure AI platform at https://ai.azure.com/ and navigate to the deployments section. Click "Deploy base model" and select your desired model (e.g., GPT-4). Complete the model deployment process.
Open the Team-GPT application and go to "Models and Credentials." Click "Add Model" and select "Azure OpenAI" as the service.
For the connection setup, enter the deployment name, which should match your selected model name. Copy the instance endpoint URL (only up to ".com/") and locate your API key. Paste both the endpoint and API key in the appropriate fields, then save the configuration.
To test your setup, go to your models section and select the configured Azure OpenAI model. Send a test message to verify that you receive an output.
Note: Ensure all credentials are correctly copied and that you have proper access permissions before proceeding with the setup.
- Part 1: Creating an Azure OpenAI Instance
- Part 2: Deploying and Configuring in Team-GPT