LLM Module API Reference

The empire_chain.llms module provides interfaces to various Language Model providers.

OpenAILLM

from empire_chain.llms import OpenAILLM

Class for interacting with OpenAI's language models.

Constructor

OpenAILLM(model_name: str = "gpt-4")

Parameters: - model_name (str): The OpenAI model to use. Options include: - "gpt-4" - "gpt-3.5-turbo" - "gpt-4-turbo"

Methods

generate()

def generate(self, prompt: str) -> str

Generate text based on a prompt.

Parameters: - prompt (str): The input prompt for text generation

Returns: - str: The generated text response

Example:

llm = OpenAILLM("gpt-4")
response = llm.generate("What is artificial intelligence?")

AnthropicLLM

from empire_chain.llms import AnthropicLLM

Class for interacting with Anthropic's Claude models.

Constructor

AnthropicLLM(model_name: str = "claude-3-sonnet")

Parameters: - model_name (str): The Anthropic model to use. Options include: - "claude-3-sonnet" - "claude-3-opus" - "claude-3-haiku"

Methods

generate()

def generate(self, prompt: str) -> str

Generate text using Claude.

Parameters: - prompt (str): The input prompt for text generation

Returns: - str: The generated text response

Example:

llm = AnthropicLLM("claude-3-sonnet")
response = llm.generate("Explain quantum computing")

GroqLLM

from empire_chain.llms import GroqLLM

Class for interacting with Groq's language models.

Constructor

GroqLLM(model_name: str = "mixtral-8x7b")

Parameters: - model_name (str): The Groq model to use. Options include: - "mixtral-8x7b" - "llama2-70b"

Methods

generate()

def generate(self, prompt: str) -> str

Generate text using Groq.

Parameters: - prompt (str): The input prompt for text generation

Returns: - str: The generated text response

Example:

llm = GroqLLM("mixtral-8x7b")
response = llm.generate("Write a poem about AI")

GeminiLLM

from empire_chain.llms import GeminiLLM

Class for interacting with Google's Gemini language models.

Constructor

GeminiLLM(model_name: str = "gemini-1.5-flash")

Parameters: - model_name (str): The Gemini model to use. Options include: - "gemini-1.5-flash" - "gemini-1.5-pro"

Methods

generate()

def generate(self, prompt: str) -> str

Generate text using Gemini.

Parameters: - prompt (str): The input prompt for text generation

Returns: - str: The generated text response

Example:

llm = GeminiLLM("gemini-1.5-flash")
response = llm.generate("Who is Lionel Messi?")

Common Features

All LLM classes share these common features:

Error Handling

try:
    llm = OpenAILLM()
    response = llm.generate("prompt")
except Exception as e:
    print(f"Error: {e}")

Environment Variables

Required environment variables: - OpenAI: OPENAI_API_KEY - Anthropic: ANTHROPIC_API_KEY - Groq: GROQ_API_KEY - Gemini: GEMINI_API_KEY

Best Practices

1. Model Selection

   # For complex reasoning
   llm = OpenAILLM("gpt-4")
   
   # For coding related tasks
   llm = AnthropicLLM("claude-3-opus")
   
   # For faster responses
   llm = GroqLLM("mixtral-8x7b")
   
   # For longer context
   llm = GeminiLLM("gemini-1.5-flash")
   

2. Error Handling

   try:
       response = llm.generate(prompt)
   except Exception as e:
       # Handle specific error types
       pass
   

3. Environment Setup

   from dotenv import load_dotenv
   load_dotenv()  # Load API keys from .env