Update app_v3.py

app_v3.py

@@ -33,8 +33,8 @@ model_basename = "Jackson2-4bit-128g-GPTQ"
 
 use_triton = False
 
-if torch.cuda.is_available():
-    torch.cuda.empty_cache()
+#if torch.cuda.is_available():
+#    torch.cuda.empty_cache()
     
 #tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True, legacy=False)
 tokenizer = AutoTokenizer.from_pretrained(pretrained_model_dir, use_fast=True)