How to call a chat model in LangChain

Chat models sit at the model boundary of a LangChain application, whether the next layer is an agent, retriever, or custom service. A direct call keeps that boundary visible before additional orchestration hides provider, credential, or message-format errors.

The current Python API can initialize standalone chat models with init_chat_model from langchain.chat_models. Provider packages such as langchain-openai supply the concrete integration, and a provider-prefixed model string such as openai:gpt-5-nano selects the provider without changing the call shape.

The script asks for a fixed response, prints the returned message class, and prints response.text. Use a throwaway prompt while confirming credentials because real provider calls can bill the account and missing or invalid keys fail before LangChain returns an AIMessage.

1. Open an activated Python project environment.

   Related: How to create a virtual environment for LangChain

2. Install LangChain and the OpenAI provider integration.

   $ python3 -m pip install --upgrade langchain langchain-openai

   langchain-openai provides the concrete OpenAI chat model implementation used by the provider-prefixed model string.

3. Set the provider API key in the shell that will run the script.

   $ export OPENAI_API_KEY="sk-..."

   Use a real key only in your local shell, secret manager, or CI secret store. Do not paste provider keys into source files or saved transcripts.

4. Create the chat model call script.

   $ cat > call_chat_model.py <<'PY'
   import os
   
   from langchain.chat_models import init_chat_model
   
   
   model_name = os.environ.get("LANGCHAIN_CHAT_MODEL", "openai:gpt-5-nano")
   
   model = init_chat_model(
       model_name,
       temperature=0,
       max_retries=0,
   )
   
   response = model.invoke(
       "Reply with exactly: LangChain chat model call succeeded."
   )
   
   print(type(response).__name__)
   print(response.text)
   PY

   Set LANGCHAIN_CHAT_MODEL to another provider-prefixed model after installing that provider package. Set OPENAI_API_BASE only when your environment routes OpenAI-compatible requests through a gateway.
   Related: How to set an OpenAI-compatible base URL in LangChain

5. Run the script.

   $ python3 call_chat_model.py
   AIMessage
   LangChain chat model call succeeded.

   AIMessage confirms that the chat model interface returned a LangChain message object. The second line confirms that the response text is available through response.text.

6. Remove the temporary script after the smoke test.

   $ rm call_chat_model.py