Skip to main content

LLM Configuration

LLM configuration system for managing language model settings, parameters, and optimization.

Overview​

The LLM configuration system provides comprehensive management of language model settings, including model selection, parameter tuning, and optimization strategies.

Core Features​

  • Model Selection: Choose from multiple LLM providers
  • Parameter Tuning: Configure temperature, max tokens, and other parameters
  • Optimization: Automatic parameter optimization
  • Cost Management: Track and optimize costs
  • Performance Monitoring: Monitor model performance

Usage Examples​

Basic LLM Configuration​

from recoagent.llm.config import LLMConfigManager

# Create configuration manager
config_manager = LLMConfigManager()

# Create LLM configuration
llm_config = config_manager.create_config(
    provider="openai",
    model="gpt-4",
    temperature=0.7,
    max_tokens=1000
)

# Use configuration
llm = config_manager.get_llm(llm_config)
response = llm.invoke("Hello, world!")

Advanced Configuration​

# Create advanced configuration with optimization
advanced_config = config_manager.create_advanced_config(
    provider="anthropic",
    model="claude-3",
    parameters={
        "temperature": 0.5,
        "max_tokens": 2000,
        "top_p": 0.9,
        "frequency_penalty": 0.1
    },
    optimization={
        "auto_tune": True,
        "cost_optimization": True,
        "performance_monitoring": True
    }
)

API Reference​

LLMConfigManager Methods​

create_config(provider: str, model: str, **kwargs) -> LLMConfig​

Create LLM configuration

Parameters:

  • provider (str): LLM provider name
  • model (str): Model name
  • **kwargs: Additional configuration parameters

Returns: LLMConfig object

get_llm(config: LLMConfig) -> BaseLanguageModel​

Get configured LLM instance

Parameters:

  • config (LLMConfig): LLM configuration

Returns: Configured LLM instance

See Also​