Skip to main content

​
Custom LLM Adapter

Create a custom LLM adapter: 
from langchat.adapters.services.openai_service import OpenAILLMService

class CustomLLMAdapter(OpenAILLMService):
    def invoke(self, messages, **kwargs):
        # Custom preprocessing
        processed_messages = self.preprocess(messages)
        
        # Call parent method
        response = super().invoke(processed_messages, **kwargs)
        
        # Custom postprocessing
        return self.postprocess(response)
    
    def preprocess(self, messages):
        # Custom preprocessing logic
        return messages
    
    def postprocess(self, response):
        # Custom postprocessing logic
        return response

​
Custom Vector Adapter

Create a custom vector adapter: 
from langchat.adapters.vector_db.pinecone_adapter import PineconeVectorAdapter

class CustomVectorAdapter(PineconeVectorAdapter):
    def get_retriever(self, k=5):
        retriever = super().get_retriever(k=k)
        # Custom retriever logic
        return retriever

​
Integration

Integrate custom adapters: 
from langchat.core.engine import LangChatEngine

class CustomLangChatEngine(LangChatEngine):
    def _initialize_adapters(self):
        # Use custom adapters
        self.llm = CustomLLMAdapter(...)
        self.vector_adapter = CustomVectorAdapter(...)
        # ... other adapters