​
Using Azure OpenAI in LobeVidol

Using Azure OpenAI in LobeVidol This document will guide you on how to use the Azure OpenAI service in LobeVidol. Azure OpenAI is an enterprise-level AI service provided by Microsoft, offering powerful models similar to OpenAI but with enhanced security and reliability guarantees.

​
Prerequisites

Before you begin, please ensure that:
  1. You have an Azure account.
  2. You have obtained access to the Azure OpenAI service.
  3. You have deployed at least one AI model.

​
Detailed Steps

​
1. Obtain Azure OpenAI API Key

First, you need to obtain the API key and endpoint for Azure OpenAI:
  1. Visit the Azure OpenAI Service Portal.
  2. If you haven’t registered yet, complete the registration process.
Register Azure OpenAI Account
  1. Go to the Deployments page and create a new deployment using your chosen model.
    • It is recommended to use the gpt-4 or gpt-35-turbo model.
    • Note down the deployment name, as it will be needed for later configuration.
Select Model to Create New Deployment
  1. Navigate to the Chat page and click View Code to obtain:
    • API endpoint URL
    • API key
Go to Chat Page Obtain Endpoint and Key

​
2. Configure in LobeVidol

Once you have the necessary information, follow these steps to complete the configuration in LobeVidol:
  1. Open LobeVidol and go to the Settings page.
  2. Find the Language Model section.
  3. Select the Azure OpenAI tab.
  4. Fill in the following information:
    • API key
    • API endpoint
    • Deployment name (optional)
Enter API Configuration
  1. Choose the Azure OpenAI model you want to use.
  2. Click save to complete the configuration.
Select Model and Start Conversation

​
Notes

Using the Azure OpenAI service incurs costs. Please refer to the Azure OpenAI Pricing Page for specific pricing details.
  • Ensure your Azure subscription has sufficient quota.
  • Regularly check API usage to avoid exceeding your budget.
  • Keep your API key secure and do not disclose it to others.

​
Frequently Asked Questions

​
What should I do if the API connection fails?

  1. Check if the API key is entered correctly.
  2. Confirm that the endpoint URL format is correct.
  3. Verify that the deployment name matches what is in the Azure portal.

​
How can I switch between different models?

In the model selection dropdown menu on the settings page, you can switch to other deployed Azure OpenAI models at any time.

​
How is billing handled?

Azure OpenAI uses a pay-as-you-go billing model, with specific rates depending on:
  • The selected model
  • The number of API calls
  • The number of tokens processed

​
More Resources