The Azure OpenAI Vision Connector enables developers to seamlessly connect and leverage the advanced artificial intelligence capabilities of Azure OpenAI.
The component allows to use the following operations:

• Get Chat Completion
• Get Image Description

Get Chat Completion
Create completions for chat messages with GPT-4 models with Vision.

Get Image Description
Use GPT-4 with vision to retrieve information from an image.

To use this connector it is necessary to have a Microsoft Azure account with the service Azure OpenAI activated.

Using the Azure OpenAI service will have cost that will be billed in your Azure Account. Check the Azure OpenAI pricing page.

Instructions

Authentication is done using API Key authentication that can be found in the Azure Portal under Resource Management > Keys and Endpoint.

In order to use the connector it is necessary to deploy models using the Azure AI Studio, specifically the model gpt-4, turbo-2024-04-09, check the region availability.

This connector requires these configurations:

Settings

ApiKey - API Key configured in Azure Portal for the Open AI Service

ApiVersion - The API version to use for this operation. This follows the YYYY-MM-DD format. Default is 2023-12-01-preview.

AzureOpenAIEndpoint - The endpoint from the Azure OpenAI Endpoint.

ChatCompletionDeploymentId - The deployment name you chose when you deployed the model for Chat Completions.