MCP Multi-Agent Orchestration Server

""" Factory for creating LLM service instances based on configuration. """ import os import logging from typing import Optional from .llm_service import LLMService from .ollama_service import OllamaService from .openai_service import OpenAIService logger = logging.getLogger(__name__) def create_llm_service( provider: Optional[str] = None, **kwargs ) -> LLMService: """ Create an LLM service instance based on provider. Args: provider: LLM provider name ("ollama" or "openai"). Defaults to LLM_PROVIDER env var. **kwargs: Additional arguments passed to the service constructor Returns: LLMService instance """ provider = provider or os.getenv("LLM_PROVIDER", "ollama").lower() if provider == "ollama": return OllamaService( base_url=kwargs.get("base_url") or os.getenv("OLLAMA_BASE_URL", "http://localhost:11434"), default_model=kwargs.get("default_model") or os.getenv("OLLAMA_MODEL", "llama3:latest") ) elif provider == "openai": api_key = kwargs.get("api_key") or os.getenv("OPENAI_API_KEY") if not api_key: raise ValueError("OpenAI API key is required. Set OPENAI_API_KEY environment variable.") return OpenAIService( api_key=api_key, default_model=kwargs.get("default_model") or os.getenv("OPENAI_MODEL", "gpt-4") ) elif provider in ["anthropic", "google"]: # These providers are listed in /api/providers but not yet implemented raise ValueError(f"Provider '{provider}' is not yet implemented. Supported providers: ollama, openai") else: raise ValueError(f"Unsupported LLM provider: {provider}. Supported providers: ollama, openai")