Chatting with Llama 3
A data analytics company is developing an internal AI assistant to help employees quickly access technical information. As part of this project, they want to integrate Llama 3, and have asked you to test it.
To get started, you initialize the Llama 3 model locally and generate a response to a question.
The model path is already stored as a variable named llama_path
.
This exercise is part of the course
Working with Llama 3
Exercise instructions
- Import the
Llama
class fromllama_cpp
. - Initialize an instance of
Llama
, specifyingllama_path
as the path to the model file.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Import the Llama class
from llama_cpp import ____
# Initialize the Llama model
llm = ____(model_path=____)
question = "What is the most used database for data storage?"
response = llm(question)
print(response)