Update app.py

app.py

@@ -46,10 +46,12 @@ def load_model():
 
     # ✅ Configure BitsAndBytes to use CPU first
     quantization_config = BitsAndBytesConfig(
-        load_in_4bit=True,
-        bnb_4bit_compute_dtype=torch.float16,
-        bnb_4bit_use_double_quant=True,
-        bnb_4bit_quant_type="nf4"
+        load_in_8bit=True,  # ✅ Uses 8-bit instead of 4-bit
+        device_map={"": "cpu"},
+        # load_in_4bit=True,
+        # bnb_4bit_compute_dtype=torch.float16,
+        # bnb_4bit_use_double_quant=True,
+        # bnb_4bit_quant_type="nf4"
     )
     
     model = AutoModelForCausalLM.from_pretrained(