Update app.py

app.py

@@ -6,12 +6,14 @@ from llama_cpp import Llama
 llm_name = "MuntasirHossain/Meta-Llama-3-8B-OpenOrca-GGUF"
 llm_path = os.path.basename(llm_name)
 
+gguf_model = "Q4_K_M" # "Q6_K.gguf" 
+
 # download gguf model
 def download_llms(llm_name):
     """Download GGUF model"""
     download_url = ""
     print("Downloading " + llm_name)
-    download_url = "https://huggingface.co/MuntasirHossain/Meta-Llama-3-8B-OpenOrca-GGUF/resolve/main/Q4_K_M.gguf" # Q6_K
+    download_url = f"https://huggingface.co/MuntasirHossain/Meta-Llama-3-8B-OpenOrca-GGUF/resolve/main/{gguf_model}"
 
     if not os.path.exists("model"):
         os.makedirs("model")
@@ -32,15 +34,13 @@ def download_llms(llm_name):
             print("Download completed")
         else:
             print(f"Model download completed {response.status_code}")
-    return llm_temp_file_path
 
 # define model pipeline with llama-cpp
 def initialize_llm(llm_model): 
     model_path = ""
     if llm_model == llm_name:
-        # model_path = "model/Q4_K_M.gguf"
+        model_path = f"model/{gguf_model}"
         download_llms(llm_model)
-        model_path=llm_temp_file_path
     llm = Llama(
         model_path=model_path,
         n_ctx=1024, # input text context length, 0 = from model