Update app.py

app.py

@@ -35,7 +35,7 @@ model.config.sliding_window = 4096
 model.eval()
 
 
-@spaces.GPU(duration=120)
+@spaces.GPU(duration=720)
 def generate(
     message: str,
     chat_history: list[tuple[str, str]],
@@ -46,7 +46,7 @@ def generate(
     repetition_penalty: float = 1.2,
 ) -> Iterator[str]:
     if chat_history != []:
-        yield "Sorry, I am an instruction-tuned model and currently do not support chatting. Please try clearing the chat history or refreshing the page to ask a new question."
+        return "Sorry, I am an instruction-tuned model and currently do not support chatting. Please try clearing the chat history or refreshing the page to ask a new question."
 
     tokenized_example = tokenizer(message, return_tensors='pt', max_length=MAX_INPUT_TOKEN_LENGTH, truncation=True)
     input_ids = tokenized_example.input_ids
@@ -70,9 +70,10 @@ def generate(
     outputs = []
     for text in streamer:
         outputs.append(text)
-        to_yield = "".join(outputs)
-        to_yield += "\n\nI have now attempted to solve the optimization modeling task! Please try executing the code in your environment, making sure it is equipped with `coptpy`."
-        yield to_yield
+        yield "".join(outputs)
+
+    outputs.append("\n\nI have now attempted to solve the optimization modeling task! Please try executing the code in your environment, making sure it is equipped with `coptpy`.")
+    yield "".join(outputs)
 
 
 chat_interface = gr.ChatInterface(