Update app.py

app.py

@@ -27,12 +27,8 @@ def ask(text):
     device_map="cuda")
 
   inputs = tokenizer.apply_chat_template(text, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda")
-
-  input_length = inputs.input_ids.shape[1]
   outputs = model.generate(**inputs, generation_config=generation_config,
                            return_dict_in_generate=True)
-
-  tokens = outputs.sequences[0, input_length:]
-  return tokenizer.decode(tokens)
+  return tokenizer.decode(outputs[0], skip_special_tokens=True)
 iface = gr.Interface(fn=ask, inputs="text", outputs="text")
 iface.launch()