Update app.py

app.py

@@ -7,6 +7,7 @@ import random
 import string
 from time import sleep
 import tiktoken
+from huggingface_hub import hf_hub_download
 
 # for counting the tokens in the prompt and in the result
 #context_count = len(encoding.encode(yourtext))
@@ -30,6 +31,15 @@ if "messages" not in st.session_state:
 if "repeat" not in st.session_state:
     st.session_state.repeat = 1.35
 
+      
+
+if "modelfile" not in st.session_state:
+    modelfile = hf_hub_download(
+        repo_id=os.environ.get("REPO_ID", "bartowski/gemma-2-2b-it-GGUF"),
+        filename=os.environ.get("MODEL_FILE", "gemma-2-2b-it-Q5_K_M.gguf"),
+    )
+    st.session_state.modelfile = modelfile
+    
 if "temperature" not in st.session_state:
     st.session_state.temperature = 0.1
 
@@ -64,9 +74,8 @@ def genRANstring(n):
 def create_chat():   
 # Set HF API token  and HF repo
     from llama_cpp import Llama
-    client = Llama.from_pretrained(
-                repo_id="bartowski/gemma-2-2b-it-GGUF",
-                filename="gemma-2-2b-it-Q5_K_M.gguf",
+    client = Llama(
+                model_path=st.session_state.modelfile,
                 temperature=0.24,
                 n_ctx=nCTX,
                 max_tokens=600,