This folder contains aa set of basic chatbot examples that are functionally identical, but using different LLMs:
| LLM. | Project | Notes |
|---|---|---|
| OpenAI/Ollama. | ./openai-ollama | This works with any OpenAI SDK compatible LLM, including Ollama running locally |
| Azure Inference | ./azure-inference | This works with models deployed to Azure AI foundry via the Azure Inference API |
| Anthropic | ./anthropic |
Navigate to the different folders for instructions on how to set them up and run the projects.
These example Python projects show how to create a basic terminal-based chatbot to interact with an LLM, with all interactions logged to Galileo. You can use this to:
- Interact with a choice of LLMs
- See traces logged to Galileo
- Measure the correctness of the responses
This project also has a unit test that runs the chatbot using the Galileo experiments functionality to unit test your system prompt and model combination.