Update app.py

app.py

@@ -65,7 +65,7 @@ from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
 # repo_id = "meta-llama/Llama-3.1-8B-Instruct"
 repo_id = "Qwen/Qwen2.5-7B-Instruct"
 
-llm_endpoint = HuggingFaceEndpoint(
+llm = HuggingFaceEndpoint(
     repo_id=repo_id,
     task="text-generation",
     max_new_tokens=512,
@@ -74,7 +74,7 @@ llm_endpoint = HuggingFaceEndpoint(
 
 # 2. ChatHuggingFace will correctly route this to the /v1/chat/completions 
 # endpoint which the Llama-3.1-8B-Instruct model supports.
-llm = ChatHuggingFace(llm=llm_endpoint)
+# llm = ChatHuggingFace(llm=llm_endpoint)
 # -------------------------
 # Document loader
 # -------------------------