AzureChatOpenAI
Azure OpenAI Service provides REST API access to OpenAI's powerful language models including the GPT-4, GPT-3.5-Turbo, and Embeddings model series. These models can be easily adapted to your specific task including but not limited to content generation, summarization, semantic search, and natural language to code translation. Users can access the service through REST APIs, Python SDK, or a web-based interface in the Azure OpenAI Studio.
This notebook goes over how to connect to an Azure-hosted OpenAI endpoint. First, we need to install the langchain-openai
package.
%pip install -qU langchain-openai
Next, let's set some environment variables to help us connect to the Azure OpenAI service. You can find these values in the Azure portal.
import os
os.environ["AZURE_OPENAI_API_KEY"] = "..."
os.environ["AZURE_OPENAI_ENDPOINT"] = "https://<your-endpoint>.openai.azure.com/"
os.environ["AZURE_OPENAI_API_VERSION"] = "2023-06-01-preview"
os.environ["AZURE_OPENAI_CHAT_DEPLOYMENT_NAME"] = "chat"
Next, let's construct our model and chat with it:
from langchain_core.messages import HumanMessage
from langchain_openai import AzureChatOpenAI
model = AzureChatOpenAI(
openai_api_version=os.environ["AZURE_OPENAI_API_VERSION"],
azure_deployment=os.environ["AZURE_OPENAI_CHAT_DEPLOYMENT_NAME"],
)
message = HumanMessage(
content="Translate this sentence from English to French. I love programming."
)
model.invoke([message])
AIMessage(content="J'adore programmer.", response_metadata={'token_usage': {'completion_tokens': 6, 'prompt_tokens': 19, 'total_tokens': 25}, 'model_name': 'gpt-35-turbo', 'system_fingerprint': None, 'prompt_filter_results': [{'prompt_index': 0, 'content_filter_results': {'hate': {'filtered': False, 'severity': 'safe'}, 'self_harm': {'filtered': False, 'severity': 'safe'}, 'sexual': {'filtered': False, 'severity': 'safe'}, 'violence': {'filtered': False, 'severity': 'safe'}}}], 'finish_reason': 'stop', 'logprobs': None, 'content_filter_results': {'hate': {'filtered': False, 'severity': 'safe'}, 'self_harm': {'filtered': False, 'severity': 'safe'}, 'sexual': {'filtered': False, 'severity': 'safe'}, 'violence': {'filtered': False, 'severity': 'safe'}}}, id='run-25ed88db-38f2-4b0c-a943-a03f217711a9-0')
Model Version
Azure OpenAI responses contain model
property, which is name of the model used to generate the response. However unlike native OpenAI responses, it does not contain the version of the model, which is set on the deployment in Azure. This makes it tricky to know which version of the model was used to generate the response, which as result can lead to e.g. wrong total cost calculation with OpenAICallbackHandler
.
To solve this problem, you can pass model_version
parameter to AzureChatOpenAI
class, which will be added to the model name in the llm output. This way you can easily distinguish between different versions of the model.
from langchain_community.callbacks import get_openai_callback
model = AzureChatOpenAI(
openai_api_version=os.environ["AZURE_OPENAI_API_VERSION"],
azure_deployment=os.environ[
"AZURE_OPENAI_CHAT_DEPLOYMENT_NAME"
], # in Azure, this deployment has version 0613 - input and output tokens are counted separately
)
with get_openai_callback() as cb:
model.invoke([message])
print(
f"Total Cost (USD): ${format(cb.total_cost, '.6f')}"
) # without specifying the model version, flat-rate 0.002 USD per 1k input and output tokens is used
Total Cost (USD): $0.000041
We can provide the model version to AzureChatOpenAI
constructor. It will get appended to the model name returned by Azure OpenAI and cost will be counted correctly.
model0301 = AzureChatOpenAI(
openai_api_version=os.environ["AZURE_OPENAI_API_VERSION"],
azure_deployment=os.environ["AZURE_OPENAI_CHAT_DEPLOYMENT_NAME"],
model_version="0301",
)
with get_openai_callback() as cb:
model0301.invoke([message])
print(f"Total Cost (USD): ${format(cb.total_cost, '.6f')}")
Total Cost (USD): $0.000044