Fix tokenizer loading issue with use_fast parameter

app.py

@@ -8,11 +8,17 @@ DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
 
 # Initialize model and tokenizer
 print(f"Loading model {MODEL_ID}...")
-tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
+try:
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, use_fast=False)
+except Exception as e:
+    print(f"Failed to load tokenizer with use_fast=False, trying with use_fast=True: {e}")
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, use_fast=True, trust_remote_code=True)
+
 model = AutoModelForCausalLM.from_pretrained(
     MODEL_ID,
     torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
-    device_map="auto" if torch.cuda.is_available() else None
+    device_map="auto" if torch.cuda.is_available() else None,
+    trust_remote_code=True
 )
 
 if not torch.cuda.is_available():