Update app.py

app.py

@@ -18,6 +18,11 @@ pipe = ov_genai.LLMPipeline(model_path, device)
 tokenizer = pipe.get_tokenizer()
 config = ov_genai.GenerationConfig(max_new_tokens=4096)
 
+def streamer(subword):
+    global output_buffer
+    output_buffer += subword
+    print(subword, end='', flush=True)
+    
 # 推理函式
 def generate_response(prompt, model_name):
     global pipe, tokenizer