Update app.py

app.py

@@ -78,7 +78,7 @@ def stream_chat(
         tokenizer = tokenizer2
 
     input_text=tokenizer.apply_chat_template(conversation, tokenize=False)
-    inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)
+    inputs = tokenizer.encode(input_text, add_generation_prompt=True, return_tensors="pt").to(device)
     streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)
     
     generate_kwargs = dict(