Python AI LLM example using langchain

.python-version

3.13

example-openai-api-call.py

import requests

API_KEY = "EMPTY"
BASE_URL = "http://localhost:12345/engines/llama.cpp/v1"
URL = f"{BASE_URL}/chat/completions"
# MODEL = "ai/gemma3"
MODEL = "ai/gemma3n"

headers = {
    "Authorization": f"Bearer {API_KEY}",
}


def main():
    system_message = {
        "role": "system",
        "content": (
            "You are a helpful assistant. "
            "Answer questions clearly an concisely. "
            "If you don't know the answer, you have to say so. "
            "Otherwise provide concise answer, don't add any extra info that user didn't ask for."
        ),
    }
    user_message = {
        "role": "user",
        # "content": "Hello, who are you?",
        # "content": "What is the capital of Russia?",
        # "content": "What is the capital of Russia? Also give one the most important fact about the capital.",
        "content": (
            "Context: square is red, triangle is yellow, hexagon is blue, circle is green\n"
            "Question: what color is hex?"
            # "Question: what color is circle?"
        ),
    }
    data = {
        "model": MODEL,
        "messages": [
            system_message,
            user_message,
        ],
    }

    response = requests.post(URL, json=data, headers=headers)
    if response.ok:
        # pprint(response.json())
        print(response.json()["choices"][0]["message"]["content"])
    else:
        print(response)
        print(response.text)


if __name__ == "__main__":
    main()

example-openai-langchain.ipynb

pyproject.toml

[project]
name = "075-ai-llm-how-to"
version = "0.1.0"
description = "Add your description here"
readme = "README.md"
requires-python = ">=3.13"
dependencies = [
    "langchain>=0.3.27",
    "langchain-openai>=0.3.33",
    "notebook>=7.4.5",
    "requests>=2.32.5",
]