Update app.py

app.py

@@ -6,7 +6,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
 tokenizer = AutoTokenizer.from_pretrained("ping98k/typhoon-7b-rag-instruct-th")
 model = AutoModelForCausalLM.from_pretrained("ping98k/typhoon-7b-rag-instruct-th", device_map={"": 0})
 
-@spaces.GPU(duration=120)
+
 '''def response(instruction, history, inputText):
     inp = f"""### Instruction:
 {instruction}
@@ -40,6 +40,7 @@ gr.ChatInterface(
     ],
 ).launch()'''
 
+@spaces.GPU(duration=120)
 def response(inp):
     input_ids = tokenizer(inp, return_tensors='pt').to("cuda")
     beam_output = model.generate(**input_ids, max_new_tokens=300)