Update app.py

app.py

@@ -193,7 +193,7 @@ parser.add_argument(
 parser.add_argument("--base-model-path", type=str, default="lmsys/vicuna-7b-v1.3",
                     help="path of basemodel, huggingface project or local path")
 parser.add_argument(
-    "--load-in-8bit", action="store_flase", help="Use 8-bit quantization"
+    "--load-in-8bit", action="store_true", help="Use 8-bit quantization"
 )
 parser.add_argument(
     "--load-in-4bit", action="store_true", help="Use 4-bit quantization"
@@ -213,7 +213,7 @@ model = EaModel.from_pretrained(
     torch_dtype=torch.float16,
     low_cpu_mem_usage=True,
     load_in_4bit=args.load_in_4bit,
-    load_in_8bit=args.load_in_8bit,
+    load_in_8bit=True,
     device_map="auto"
 )
 model.eval()