Azure AI Inference Chat Streaming in Python

This page was automatically generated by AI; not yet reviewed for accuracy...

The content and code samples on this page were generated by using the ai CLI with customized prompts in this repository.

It's cool, but, it's experimental. 😁

Please review the content and code before using it in your application.

This sample demonstrates how to use the Azure AI Inference Chat API with streaming in a Python application.

main.py
azureml_chat_completions_streaming.py

How to generate this sample

Command

ai dev new az-inference-chat-streaming --python

Output

AI - Azure AI CLI, Version 1.0.0
Copyright (c) 2024 Microsoft Corporation. All Rights Reserved.

This PUBLIC PREVIEW version may change at any time.
See: https://aka.ms/azure-ai-cli-public-preview

Generating 'az-inference-chat-streaming' in 'az-inference-chat-streaming-py' (3 files)...

azureml_chat_completions_streaming.py
main.py
requirements.txt

Generating 'az-inference-chat-streaming' in 'az-inference-chat-streaming-py' (3 files)... DONE!

main.py

STEP 1: Read the configuration settings from environment variables:

main.py

chat_api_key = os.getenv("AZURE_AI_CHAT_API_KEY", '<insert your Azure AI Inference API key here>')
chat_endpoint = os.getenv("AZURE_AI_CHAT_ENDPOINT", '<insert your Azure AI Inference endpoint here>')
chat_model = os.getenv('AZURE_AI_CHAT_MODEL', '')
chat_system_prompt = os.getenv('SYSTEM_PROMPT', 'You are a helpful AI assistant.')

STEP 2: Initialize the helper class with the configuration settings:

main.py

chat = AzureAIInferenceChatCompletionsStreaming(chat_endpoint, chat_api_key, chat_model, chat_system_prompt)

STEP 3: Obtain user input, use the helper class to get the assistant's response, and display responses as they are received:

main.py

while True:
{
    user_input = input('User: ')
    if user_input == 'exit' or user_input == '':
        break

    print('\nAssistant: ', end='')
    response = chat.get_chat_completions(user_input, lambda content: print(content, end=''))
    print('\n')
}

azureml_chat_completions_streaming.py

STEP 1: Create the client and initialize chat message history with a system message:

azureml_chat_completions_streaming.py

class AzureAIInferenceChatCompletionsStreaming:
    def __init__(self, chat_endpoint, chat_api_key, chat_model, chat_system_prompt):
        self.chat_system_prompt = chat_system_prompt
        self.chat_model = chat_model
        self.client = ChatCompletionsClient(endpoint=chat_endpoint, credential=AzureKeyCredential(chat_api_key))
        self.clear_conversation()

    def clear_conversation(self):
        self.messages = [
            SystemMessage(content=self.chat_system_prompt)
        ];

STEP 2: When the user provides input, add the user message to the chat message history:

azureml_chat_completions_streaming.py

    def get_chat_completions(self, user_input, callback):
        self.messages.append(UserMessage(content=user_input))

STEP 3: Send the chat message history to the streaming Azure AI Chat API and process each update:

azureml_chat_completions_streaming.py

        complete_content = ''
        response = self.client.complete(
            messages=self.messages,
            model=self.chat_model,
            stream=True,
        )

        for update in response:

            if update.choices is None or len(update.choices) == 0: 
                continue

            content = update.choices[0].delta.content or ""
            if content is None: continue

            complete_content += content
            callback(content)

STEP 4: For each non-empty update, accumulate the response, and invoke the callback for the update:

azureml_chat_completions_streaming.py

        self.messages.append(AssistantMessage(content=complete_content))
        return complete_content