The Microsoft Azure OpenAI package provides REST API access to Azure-hosted language models, including the GPT-4 series and Embeddings models. Use this package to build generative AI-based automations that process natural language, generate content, and respond to complex queries within Automation Anywhere.
Microsoft Azure OpenAI service provides REST API access to language models including the GPT-3, Codex, and Embeddings model series. You can use Automation Anywhere's Microsoft Azure OpenAI package to connect with the Microsoft Azure OpenAI service REST APIs to build generative AI based automations that respond to your general natural language queries.
Overview
Microsoft Azure OpenAI service provides REST API access to Azure-hosted language models including the GPT-4 series and Embeddings models. You can use Automation Anywhere's Microsoft Azure OpenAI package to connect with the Microsoft Azure OpenAI service REST APIs to build generative AI-based automations that respond to natural language queries, generate content, and process complex language tasks.
Supported models
The Microsoft Azure OpenAI package supports the following Azure-hosted model families:
- GPT-4 series: GPT-4, GPT-4 Turbo, and GPT-4o — high-capability models for natural language understanding, content generation, code assistance, and complex reasoning tasks.
- Embeddings models: text-embedding-3-small and text-embedding-3-large — used to convert text to high-dimensional vector representations for semantic search, similarity matching, and retrieval-augmented generation (RAG) workflows.
Prerequisites
Before using the Microsoft Azure OpenAI package, ensure you have the following:
- An active Microsoft Azure subscription with Azure OpenAI access granted. Azure OpenAI access requires a separate application approval from Microsoft.
- A provisioned Azure OpenAI resource with at least one model deployment configured in the Azure portal.
- The Endpoint URL and API Key from your Azure OpenAI resource, available under Keys and Endpoint in the Azure portal.
- The Deployment name for the model you intend to use, as configured in Azure AI Studio.