Update app.py

app.py

@@ -26,7 +26,8 @@ DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
 
 # ── Model Initialization ──────────────────────────────────────────────────────
 print(f"[INFO] Loading model: {MODEL_NAME} | device: {DEVICE}")
-tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+# Force use_fast=False to avoid the sentencepiece backend error
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, use_fast=False)
 model = AutoModelForSeq2SeqLM.from_pretrained(MODEL_NAME).to(DEVICE)
 model.eval()
 print("[INFO] Model ready.")