Update app.py

app.py

@@ -29,7 +29,7 @@ def generate_response(message, history):
     conversation += f"user: {message}\nassistant: "
     
     condition = "<|quad_end|><|object_ref_end|>"
-    prompt = f"<|im_start|>{condition}{conversation} <|im_end|>"
+    prompt = f"<|im_start|>{condition}{conversation} "
     
     inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
     # Mark the prompt as a single bidirectional prefix block — see "PrefixLM mask" below.