@core-ai/azure-openai

Azure OpenAI provider package for @core-ai/core-ai. Azure v1 uses the Responses API by default and also exposes strict Chat Completions.

Installation

npm install @core-ai/core-ai @core-ai/azure-openai zod

Usage

import { generate } from '@core-ai/core-ai';
import { createAzureOpenAI } from '@core-ai/azure-openai';

const azure = createAzureOpenAI({
    apiKey: process.env.AZURE_OPENAI_API_KEY,
    endpoint: process.env.AZURE_OPENAI_ENDPOINT,
});

// Use your Azure OpenAI deployment name as the model id.
const model = azure.chatModel('gpt-5-mini-deployment');

const result = await generate({
    model,
    messages: [{ role: 'user', content: 'Hello!' }],
});

console.log(result.content);

Use Chat Completions explicitly when needed:

const model = azure.chat.chatModel('gpt-5-mini-deployment');

Authentication

Use an Azure OpenAI API key with the v1 API:

const azure = createAzureOpenAI({
    apiKey: process.env.AZURE_OPENAI_API_KEY,
    endpoint: process.env.AZURE_OPENAI_ENDPOINT,
});

Set api: 'classic' to use the classic Azure API surface:

const azure = createAzureOpenAI({
    api: 'classic',
    apiKey: process.env.AZURE_OPENAI_API_KEY,
    endpoint: process.env.AZURE_OPENAI_ENDPOINT,
    apiVersion: '2025-04-01-preview',
});

Classic mode also accepts an Entra ID token provider via azureADTokenProvider.

Model IDs

Pass your Azure OpenAI deployment name to either model factory.

const model = azure.chatModel('customer-support-gpt-5-mini');

Classic mode supports Chat Completions only, so its root chatModel() and chat.chatModel() methods are equivalent. This package does not expose embeddings or image generation.