Update backend/query_llm.py

backend/query_llm.py

@@ -106,14 +106,19 @@ def generate_hf(prompt: str, history: str, temperature: float = 0.5, max_new_tok
             return "I do not know what happened, but I couldn't understand you."
 
 def generate_qwen(formatted_prompt: str, history: str):
-    response = client.predict(
+    stream = client.predict(
                     query=formatted_prompt,
-                    history=history,
+                    history=[],
                     system="You are a helpful assistant.", 
                     api_name="/model_chat"
                 )
-    print('Response:',response)
-    return response[1][0][1]
+    print('Response:',stream)
+    output = ""
+    for response in stream:
+        output += response.token.text
+        yield output
+    return output
+    #return response[1][0][1]