Update app.py

app.py

@@ -69,7 +69,6 @@ def stream_chat(message: str, history: list, temperature: float, max_new_tokens:
     
     input_ids = tokenizer.apply_chat_template(conversation, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)
     
-    prompt_padded_len = len(input_ids[0])
 
     gen_tokens= model.generate(
         input_ids, 
@@ -78,11 +77,7 @@ def stream_chat(message: str, history: list, temperature: float, max_new_tokens:
         temperature=temperature,
     )
 
-    gen_tokens = [
-      gt[prompt_padded_len:] for gt in gen_tokens
-    ]
-
-    gen_text = tokenizer.batch_decode(gen_tokens, skip_special_tokens=True)
+    gen_text = tokenizer.decode(gen_tokens[0], skip_special_tokens=True)
 
     return gen_text