llama-cpp-agent Installation to Use AI Models Locally in Simple Way

 This video is a step-by-step easy tutorial to install llama-cpp-agent which is a tool designed to simplify interactions with LLMs. It provides an interface for chatting with LLMs, executing function calls, generating structured output.

Code:

conda create -n llamagent python=3.11 -y && conda activate llamagent

pip install --no-cache-dir llama-cpp-python==0.2.77 --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu124

pip install llama-cpp-agent

conda install jupyter -y

pip uninstall charset_normalizer -y

pip install charset_normalizer

jupyter notebook

# Import the Llama class of llama-cpp-python and the LlamaCppPythonProvider of llama-cpp-agent

from llama_cpp import Llama

from llama_cpp_agent.providers import LlamaCppPythonProvider

# Create an instance of the Llama class and load the model

llama_model = Llama(r"/home/Ubuntu/mymodels/mistral-7b-instruct-v0.2.Q6_K.gguf", n_batch=1024, n_threads=10, n_gpu_layers=40)

# Create the provider by passing the Llama class instance to the LlamaCppPythonProvider class

provider = LlamaCppPythonProvider(llama_model)

from llama_cpp_agent import LlamaCppAgent

from llama_cpp_agent import MessagesFormatterType

agent = LlamaCppAgent(provider, system_prompt="You are a helpful assistant.", predefined_messages_formatter_type=MessagesFormatterType.MISTRAL)

agent_output = agent.get_chat_response("Hello, World!")

print(f"Agent: {agent_output.strip()}")

import math

from llama_cpp_agent import FunctionCallingAgent

from llama_cpp_agent.llm_output_settings import LlmStructuredOutputSettings

from typing import Union

# Callback for receiving messages for the user.

def send_message_to_user_callback(message: str):

    print(message)

   

def calculate_a_to_the_power_b(a: Union[int, float], b: Union[int, float]):

    """

    Calculates a to the power of b

    Args:

        a: number

        b: exponent

    """

    return f"Result: {math.pow(a, b)}"

output_settings = LlmStructuredOutputSettings.from_functions([calculate_a_to_the_power_b], allow_parallel_function_calling=True)

llama_cpp_agent = LlamaCppAgent(

    provider,

    debug_output=True,

    system_prompt=f"You are an advanced AI, tasked to assist the user by calling functions in JSON format.",

    predefined_messages_formatter_type=MessagesFormatterType.CHATML,

)

user_input = "Calculate a to the power of b: a = 2, b = 3"

print(

    llama_cpp_agent.get_chat_response(

        user_input, structured_output_settings=output_settings

    )

)