Update app.py

app.py

@@ -16,7 +16,6 @@ from langchain_core.runnables import RunnableLambda
 from datetime import date
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from setup import OLMO_MODEL
-from langchain_core.callbacks import CallbackManager
 model_path = OLMO_MODEL
 
 # Environment variables
@@ -84,13 +83,9 @@ def get_chain(temperature):
     retriever = load_retriever(docstore_path,chroma_path,embeddings,child_splitter,parent_splitter)
     
     # Replace the local OLMOLLM with the Hugging Face model
-    callback_handler.on_llm_end = lambda response, *args, **kwargs: None
-    # Callbacks support token-wise streaming
-    callback_manager = CallbackManager([callback_handler])
     llm = LlamaCpp(
         model_path=str(model_path),
-        callback_manager=callback_manager,
-        temperature=0.8,
+        temperature=temperature,
         max_tokens=3000,
         verbose=False,
         echo=False