added stream

app.py

@@ -19,8 +19,10 @@ llm = AutoModelForCausalLM.from_pretrained(model_name, model_file=model_file,
                                            model_type="mistral", gpu_layers=0)
 def alternatingly_agree(message, history):
     prompt = prompt.format(message)
-    result = llm(message,max_new_tokens=50,temperature=0.7)
-    return result #stream=True
+    result = llm(message,max_new_tokens=50,temperature=0.7,stream=True)
+    for i in result:
+        yield i
+ 
  
 
 gr.ChatInterface(alternatingly_agree).launch()