LLMClient is a simple Python library for interacting with Large Language Models (LLMs). It is tailored to my own needs, but if you find it useful, you can use it too.

For now, it supports two types of LLMs:

• Ollama - https://ollama.com/
• WebChat - with a defined endpoint (it is tested with Ollama's webchat interface at http://localhost:11434/api/chat)

Both support setting a desired model, temperature and eventually a JSON format of the output. You can also use a specific system prompt (see llm_client_test_json.py for an example).

See the examples in the llm_client_test_*.py files. The most simple example is:

from llm_client.LLMClientOllama import LLMClientOllama

llm_client_ollama = LLMClientOllama()
llm_client_ollama.set_model("gemma3:12b")
llm_client_ollama.set_temperature(1)

response, llm_role = llm_client_ollama.call_llm("What is the meaning of life and everything?")
print(f"Response from {llm_role}: {response}")

This project uses the following dependencies:

• requests - for making HTTP requests
• json - for working with JSON data
• ollama - for working with Ollama API https://github.com/ollama/ollama-python

This project is licensed under the MIT License - see the LICENSE file for details.