Update app.py

app.py

@@ -11,13 +11,15 @@ CORS(app)
 # Model (UNGATED, FREE)
 # ----------------------------
 MODEL_NAME = "microsoft/phi-3-mini-4k-instruct"
-DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+DEVICE = "cpu"
 
-tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
 model = AutoModelForCausalLM.from_pretrained(
     MODEL_NAME,
-    torch_dtype=torch.float16 if DEVICE == "cuda" else torch.float32
-).to(DEVICE)
+    torch_dtype=torch.float32
+)
+
+model.to(DEVICE)
+)
 
 # ----------------------------
 # SYSTEM PROMPT