How to call openai chat model using langchain

To use the ChatOpenAI model from Langchain to get a response for messages, you can follow these steps:

• Install the OpenAI Python package by running pip install openai.
• Obtain an API key from OpenAI by creating an account and visiting their API key page.
• Set the API key as an environment variable by running export OPENAI_API_KEY="your-api-key", or set the key as parameter in the function (see below).
• Import the ChatOpenAI class from langchain.chat_models.
• Initialize an instance of ChatOpenAI with the API key: chat = ChatOpenAI(openai_api_key="your-api-key") ( pass api key as parameter If api key is not set in environment).
• Create a list of messages to send to the model. Messages can be of types AIMessage, HumanMessage, or SystemMessage.
• Invoke the model by calling chat.invoke(messages). This will return an AIMessage containing the model’s response.
• Alternatively, you can use chat.stream(messages) to stream the model’s response in chunks.
• You can also use chat.batch([messages]) to batch process multiple sets of messages.

Here is an example of using ChatOpenAI to get a response for messages:

from langchain.chat_models import ChatOpenAI
from langchain.schema.messages import HumanMessage, SystemMessage

chat = ChatOpenAI(openai_api_key="your-api-key")

messages = [
    SystemMessage(content="You're a helpful assistant"),
    HumanMessage(content="What is the purpose of model regularization?"),
]

response = chat.invoke(messages)
print(response.content)

This will output the model’s response to the given messages.