Update app.py

app.py

@@ -61,6 +61,8 @@ tasks = {
 }
 
 os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:30'
+os.environ['CUDA_LAUNCH_BLOCKING'] = 1
+os.environ['CUDA_CACHE_DISABLE'] = 1
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
 # Define the model and tokenizer globally