Update app.py

app.py

@@ -46,7 +46,7 @@ def inference(query):
     tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda")
     outputs = model.generate(tokenized_chat, **generation_params)
     decoded_outputs = tokenizer.batch_decode(outputs, skip_special_tokens=True)
-    assistant_response = decoded_outputs[0].split("<|im_start|>assistant\n")[-1].strip()
+    assistant_response = decoded_outputs[0][len(messages):]).split("<|im_start|>assistant\n")[-1].strip()
     response_ = assistant_response.replace('<|im_end|>', "")
     return response_
     # outputs = model.generate(tokenized_chat, **generation_params, streamer=streamer)