Messages in ChatModel Demo

In this lesson we’ll examine message objects and best practices for constructing prompts for chat-based models. The goal is to demonstrate the typical flow: import message classes, initialize a chat model, create SystemMessage and HumanMessage objects, send them to the model, and inspect the returned AI message.

Module and API names in SDKs such as LangChain change frequently. If an import or method shown here fails, consult the latest SDK documentation and adapt import paths or method names accordingly.

Prerequisites

An OpenAI-compatible API key available in your environment.
A compatible version of the chat SDK you plan to use (e.g., LangChain). If imports differ, check the package docs.

If the model call fails with authentication errors, confirm that your OPENAI_API_KEY is set and that your SDK supports the provider and model you are calling.

Environment variable

Set your OpenAI API key in the shell before running Python examples. macOS / Linux (bash/zsh):

export OPENAI_API_KEY="your_api_key_here"

Windows (PowerShell):

$env:OPENAI_API_KEY = "your_api_key_here"

You can also set the key programmatically in Python (not recommended for production):

import os
os.environ["OPENAI_API_KEY"] = "your_api_key_here"

Minimal Python example

This example shows a compact, runnable pattern using message objects and a chat model. Adjust imports and model initialization for your SDK version if necessary.

# example_chat_messages.py
from langchain.chat_models import ChatOpenAI
from langchain.schema import SystemMessage, HumanMessage
import os

# Optional: set the API key from Python (best practice is to use environment variables)
# Initialize the chat model (adjust parameters like temperature, model name as needed)
model = ChatOpenAI()

# Define the assistant persona and the user's prompt
sysmsg = "You are a Physics teacher."
humanmsg = "Explain the concept of a galaxy."

messages = [
    SystemMessage(content=sysmsg),
    HumanMessage(content=humanmsg),
]

# Invoke the model with the list of messages.
# Many LangChain chat model wrappers provide `predict_messages` which returns an AIMessage-like object.
response = model.predict_messages(messages)

# Inspect the returned AIMessage object
print(response)            # full AIMessage representation
print(response.content)    # just the textual content returned by the model

Note: If your SDK exposes a different method (e.g., generate or chat), adapt the invocation accordingly.

Example API response (illustrative)

The returned object is commonly an AIMessage-like structure. Example content (actual output varies):

AIMessage(
    content="A galaxy is a massive ensemble of stars, stellar remnants, interstellar gas, dust, dark matter, and other celestial objects bound together by gravity. Galaxies come in several shapes — spiral, elliptical, and irregular — and form the large-scale structure of the universe. The Milky Way, our home galaxy, is a spiral galaxy..."
)

Message roles — quick reference

Role	Purpose	Typical use
SystemMessage	Defines persona, global instructions, or assistant behavior	`SystemMessage(content="You are a concise technical writer.")`
HumanMessage	The user’s prompt or question	`HumanMessage(content="Explain recursion in simple terms.")`
AIMessage	The model’s response (returned by the API)	Inspect with `response.content`

Key points & troubleshooting

Keep SDK imports and method names up to date with the official docs. If from langchain.schema import SystemMessage, HumanMessage fails, consult the package docs for the correct path.
If you receive authentication or quota errors, verify that OPENAI_API_KEY is set in the environment and that your account has available quota.
For multi-turn conversations, append subsequent HumanMessage and AIMessage instances to the messages list to preserve context across turns.
Adjust model parameters (temperature, max tokens, model name) via the chat model constructor or call arguments depending on your SDK.

Introduction

Overview of LangChain

Key Components of LangChain

Interacting with LLMs

Tips, Tricks, and Resources

Introduction to LCEL

Adding Memory to LLM Apps

Performing Retrieval

Implementing Chains

Using Tools

Building Agents

Conclusion

Building Blocks of LLM Apps

Messages in ChatModel Demo

Prerequisites

Environment variable

Minimal Python example

Example API response (illustrative)

Message roles — quick reference

Key points & troubleshooting

Further reading

Watch Video

Practice Lab

​Prerequisites

​Environment variable

​Minimal Python example

​Example API response (illustrative)

​Message roles — quick reference

​Key points & troubleshooting

​Further reading

Watch Video

Practice Lab

Prerequisites

Environment variable

Minimal Python example

Example API response (illustrative)

Message roles — quick reference

Key points & troubleshooting

Further reading