Update app_v1.py

app_v1.py

@@ -3,7 +3,13 @@ from transformers import AutoTokenizer
 from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
 from huggingface_hub import snapshot_download
 import os
-import threading
+import torch
+
+# Clear up some memory
+torch.cuda.empty_cache()
+
+# Try reducing the number of threads PyTorch uses
+torch.set_num_threads(1)
 
 cwd = os.getcwd()
 cachedir = cwd + '/cache'