Swarmauri Azure OpenAI LLM

swarmauri_llm_azureopenai provides provider-native Swarmauri adapters for Azure OpenAI and Microsoft Foundry chat-completions deployments.

Features

• AzureOpenAIModel directly implements the Swarmauri LLM base contract.
• AzureOpenAIToolModel directly implements the tool-LLM base contract.
• API-key authentication with Azure's api-key header.
• Refreshable Microsoft Entra bearer-token providers kept outside serialized state.
• Sync, async, streaming, and batch workflows against the current /openai/v1 API.
• Persistent, injectable Swarmauri HTTP clients with conservative rate-limit retries.

Installation

uv add swarmauri_llm_azureopenai

pip install swarmauri_llm_azureopenai

Usage

import os

from swarmauri_llm_azureopenai import AzureOpenAIModel
from swarmauri_standard.conversations.Conversation import Conversation
from swarmauri_standard.messages.HumanMessage import HumanMessage

conversation = Conversation(messages=[HumanMessage(content="Summarize this release.")])
model = AzureOpenAIModel(
    endpoint=os.environ["AZURE_OPENAI_ENDPOINT"],
    api_key=os.environ["AZURE_OPENAI_API_KEY"],
    name="my-gpt-deployment",
)
model.predict(conversation)
print(conversation.get_last().content)

For keyless authentication, omit api_key and pass a callable as token_provider; it is invoked for every request so refreshed Entra tokens are used. Use AzureOpenAIToolModel with a Swarmauri Toolkit for function calling.

Pass an IHttpClient as http_client to integrate a custom transport or to test without network access. Pass an IHttpRetryPolicy as retry_policy to replace the default policy, which retries safe reads and rate-limited chat submissions without replaying failed POST requests for generic server errors.

See the Azure OpenAI chat REST reference and Entra authentication guide.