Update app.py

app.py

@@ -14,9 +14,8 @@ print(f"📦 Using model: {MODEL_ID}")
 print("=" * 60)
 
 # ================= LOAD MODEL =================
-@gr.cache_resource
 def load_model():
-    """Load model dengan caching"""
+    """Load model"""
     print("📥 Loading model...")
     
     try:
@@ -51,7 +50,7 @@ def load_model():
     
     return tokenizer, model
 
-# Load model
+# Load model - only once at startup
 tokenizer, model = load_model()
 
 # ================= GENERATION FUNCTION =================