LLM Manager API

This page documents the Language Model management components.

LLMManager

class LLMManager:
    """Manages Ollama language model interactions."""

    def __init__(self, model_name: str = "llama2"):
        """Initialize LLM manager with model name."""

Methods

list_models

def list_models() -> List[str]:
    """List available Ollama models."""

Returns: - List of model names

get_model

def get_model(self, model_name: str) -> LLM:
    """Get an instance of the specified model."""

Parameters: - model_name: Name of the Ollama model

Returns: - LLM instance

generate

def generate(self, prompt: str, **kwargs) -> str:
    """Generate text using the current model."""

Parameters: - prompt: Input text - **kwargs: Additional generation parameters

Returns: - Generated text

Usage Example

# Initialize manager
manager = LLMManager(model_name="llama2")

# List available models
models = manager.list_models()

# Generate text
response = manager.generate(
    prompt="Explain RAG in simple terms",
    temperature=0.7,
    max_tokens=500
)

Model Parameters

Configure model behavior with:

temperature: Creativity (0.0-1.0)
max_tokens: Response length
top_p: Nucleus sampling
frequency_penalty: Repetition control

Error Handling

The manager handles:

Model loading errors
Generation timeouts
Resource constraints
API communication issues

Best Practices

Model Selection
Match model to task
Consider resource usage
Test performance
Parameter Tuning
Adjust temperature
Control response length
Balance quality/speed ```