"""LLM provider abstraction layer.
This module provides a unified interface for interacting with different
LLM providers (Ollama, OpenAI, Gemini, etc.) through a protocol-based
architecture.
Public API:
- LLMProvider: Protocol defining provider interface
- LLMManager: Provider lifecycle management
- LLMConfig: Configuration models
- Request/Response types
- Exception hierarchy
"""
from bbsbot.llm.base import LLMProvider
from bbsbot.llm.config import GeminiConfig, LLMConfig, OllamaConfig, OpenAIConfig
from bbsbot.llm.exceptions import (
LLMAuthenticationError,
LLMConnectionError,
LLMError,
LLMInvalidResponseError,
LLMModelNotFoundError,
LLMRateLimitError,
LLMTimeoutError,
)
from bbsbot.llm.manager import LLMManager
from bbsbot.llm.types import (
ChatMessage,
ChatRequest,
ChatResponse,
CompletionRequest,
CompletionResponse,
StreamChunk,
)
__all__ = [
# Core
"LLMProvider",
"LLMManager",
# Config
"LLMConfig",
"OllamaConfig",
"OpenAIConfig",
"GeminiConfig",
# Types
"ChatMessage",
"ChatRequest",
"ChatResponse",
"CompletionRequest",
"CompletionResponse",
"StreamChunk",
# Exceptions
"LLMError",
"LLMConnectionError",
"LLMTimeoutError",
"LLMRateLimitError",
"LLMModelNotFoundError",
"LLMInvalidResponseError",
"LLMAuthenticationError",
]
