Tried

pipeline.py

@@ -26,7 +26,7 @@ from langchain_community.vectorstores import FAISS
 from langchain.chains import RetrievalQA, LLMChain
 from langchain.prompts import PromptTemplate
 from langchain.docstore.document import Document
-
+from langchain_core.caches import BaseCache
 # from langchain.callbacks.base import BaseCallbacks  # Updated import
 # from langchain.callbacks.manager import CallbackManager
 # from langchain.callbacks import StdOutCallbackHandler
@@ -192,69 +192,36 @@ CACHE_SIZE_LIMIT = 1000
 # except Exception as e:
 #     logger.error(f"Failed to initialize ChatGroq: {e}")
 #     raise RuntimeError("ChatGroq initialization failed.") from e
+# Define a simple no-op cache class
+class NoCache(BaseCache):
+    """Simple no-op cache implementation."""
+    def __init__(self):
+        pass
+
+    def lookup(self, prompt, llm_string):
+        return None
+
+    def update(self, prompt, llm_string, return_val):
+        pass
 
+    def clear(self):
+        pass
+
+# Initialize ChatGroq with cache
 try:
-        # Get API key from environment variables
     fallback_groq_api_key = os.environ.get("GROQ_API_KEY_FALLBACK", os.environ.get("GROQ_API_KEY"))
-    
     if not fallback_groq_api_key:
         logger.warning("No Groq API key found for fallback LLM")
-        
-    # Initialize the Groq client as fallback
     groq_fallback_llm = ChatGroq(
         model=GROQ_MODELS["default"],
         temperature=0.7,
         groq_api_key=fallback_groq_api_key,
-        max_tokens=2048
+        max_tokens=2048,
+        cache=NoCache()  # Set cache explicitly
     )
-    
-    # # Verify connection by checking models (optional)
-    # try:
-    #     groq_fallback_llm.list_models()
-    #     logger.info("Fallback Groq LLM initialized successfully")
-    # except Exception as e:
-    #     logger.error(f"Fallback Groq LLM connection test failed: {e}")
 except Exception as e:
-        logger.error(f"Failed to initialize fallback Groq LLM: {e}")
-
-# from langchain_core.callbacks import CallbackManager
-# from langchain_core.callbacks.base import BaseCallbackHandler
-# from langchain_core.caches import BaseCache
-
-# # Initialize ChatGroq with the callback manager
-# try:
-#     # First, ensure BaseCache is defined (if needed)
-#     class NoCache(BaseCache):
-#         """Simple no-op cache implementation."""
-#         def __init__(self): 
-#             pass
-            
-#         def lookup(self, prompt, llm_string): 
-#             return None
-            
-#         def update(self, prompt, llm_string, return_val): 
-#             pass
-            
-#         def clear(self): 
-#             pass
-    
-#     # Call model_rebuild() before initializing
-#     ChatGroq.model_rebuild()
-    
-#     # Then initialize the model
-#     fallback_groq_api_key = os.environ.get("GROQ_API_KEY_FALLBACK", os.environ.get("GROQ_API_KEY"))
-#     groq_fallback_llm = ChatGroq(
-#         model=GROQ_MODELS["default"],
-#         temperature=0.7,
-#         groq_api_key=fallback_groq_api_key,
-#         max_tokens=2048,
-#         callback_manager=CallbackManager([])  # Use CallbackManager instead of raw list
-#     )
-        
-# except Exception as e:
-#     logger.error(f"Failed to initialize ChatGroq: {e}")
-#     raise RuntimeError("ChatGroq initialization failed.") from e
-
+    logger.error(f"Failed to initialize fallback Groq LLM: {e}")
+    raise RuntimeError("ChatGroq initialization failed.") from e
 
 # -------------------------------------------------------
 # Rate-limit & Cache