remove gpu requirements.

app.py

@@ -35,12 +35,12 @@ llm = LlamaCpp(
 )
 llm_chain = prompt | llm
 
-@spaces.GPU(duration=120)
+# @spaces.GPU(duration=120)
 def greet(name):
-    logger.info("This is an info message")
     question = name
-    # print(llm_chain.invoke({"question": question}))
-    return llm_chain.invoke({"question": question})
+    response = llm_chain.invoke({"question": question})
+    logger.info(f"Response --> {response}")
+    return 
 
 demo = gr.Interface(fn=greet, inputs="text", outputs="text")
 demo.launch()