Skip to main content

Azure Mistral AI

What is Azure Mistral AI?

Azure Mistral AI is a cloud-based service provided by Microsoft that integrates Mistra's language models into Azure platforms. It allows Wordbee to incorporate Mistral AI’s advanced natural language processing features into Wordbee Translator with the reliability, security, and scalability of Microsoft's cloud infrastructure.

Configuration

To integrate your Microsoft Azure Mistral AI language model, simply go to Settings > Large Language Models (LLM) > Select. On this screen, click on Add new and choose Azure Mistral AI from the available options.

Begin by naming the system for easy reference in Wordbee Translator and set the scope for the LLM profile. The scope will define how you can use the profile in the future. To proceed with the connection, be sure to enter the required credentials.

Key

This is your authentication key for accessing Azure Mistral services.

After creating an Azure Mistral service instance in the Azure portal, you can find your key(s) under the Keys and Endpoint section of your Mistral service resource.

Base URL

The Base URL in Azure services is the primary web address used to access and interact with Azure's REST APIs.

Typically, when you create or configure a service in Azure, the Base URL is displayed in the service's dashboard or in the API management section.

To find this information, please:

  • Log into the Azure Portal (http://portal.azure.com ).

  • Navigate to the All resources or directly to AI + Machine Learning > Mistral AI service instances you've created.

  • For keys, model choices, and API versions, look into the specific Mistral AI service instance's settings and documentation within the Azure portal.

If you're new to Azure or these steps seem unfamiliar, please consult the Azure documentation for the Mistral AI service or seek assistance from someone with Azure experience.

After entering the required information, click Save to store your credentials and add the profile to the list.

JavaScript errors detected

Please note, these errors can depend on your browser setup.

If this problem persists, please contact our support.