tweak 3

app.py

@@ -6,6 +6,12 @@ from huggingface_hub import hf_hub_download
 
 # huggingface-cli download microsoft/Phi-3-mini-4k-instruct-gguf Phi-3-mini-4k-instruct-q4.gguf --local-dir .
 # huggingface-cli download LoneStriker/OpenBioLLM-Llama3-8B-GGUF --local-dir ./llama3-gguf
+# Explicitly create the cache directory if it doesn't exist
+cache_dir = os.path.expanduser("~/.cache/huggingface/hub")
+os.makedirs(cache_dir, exist_ok=True)
+
+# Use the cache directory for huggingface downloads
+os.environ["HF_HOME"] = cache_dir
 
 llm = Llama(
     # model_path="./Phi-3-mini-4k-instruct-q4.gguf",