Update app.py

app.py

@@ -5,7 +5,7 @@ from threading import Thread
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 
 # Choose any chat model with a chat template; Zephyr works well:
-MODEL_NAME = "google/gemma-3-270m"
+MODEL_NAME = "google/gemma-3-270m-it"
 
 # Load model + tokenizer
 tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, use_fast=True)