Update app.py

app.py

@@ -31,14 +31,13 @@ def load_pipeline():
 
 @spaces.GPU
 def chat_interface(inputs):
-    question = inputs["input"]
-    
+    question = inputs
     chat_history_tuples = []
     for message in chat_history:
         chat_history_tuples.append((message[0], message[1]))
     
     #result = llm_chain({"input": query, "history": chat_history_tuples})
-    result = llm_chain.invoke({"input": question, "history": chat_history})
+    result = llm_chain.invoke({"input": inputs, "history": chat_history})
     return result["response"]
 
 llm = load_pipeline()