Update app.py

app.py

@@ -18,7 +18,7 @@ OLLAMA_SERVICE_THREAD.start()
 
 print("Giving ollama serve a moment")
 time.sleep(10)
-subprocess.run("~/ollama pull gemma2", shell=True)
+
 
 
 import copy
@@ -35,11 +35,14 @@ MODEL_NAME = MODEL_ID.split("/")[-1]
 
 os.environ["HF_HUB_ENABLE_HF_TRANSFER"] = "1"
 
-@spaces.GPU()
-def gemma2():
-    return Ollama(model="gemma2:27b", request_timeout=30.0)
 
 
+llm = Ollama(model="gemma2:27b", request_timeout=60.0)
+
+
+def run_ollama():
+    subprocess.run("~/ollama pull gemma2", shell=True)
+
 TITLE = "<h1><center>Chatbox</center></h1>"
 
 DESCRIPTION = f"""
@@ -63,7 +66,7 @@ h3 {
     text-align: center;
 }
 """
-
+@spaces.GPU()
 def stream_chat(message: str, history: list, temperature: float, context_window: int, top_p: float, top_k: int, penalty: float):
     print(f'message is - {message}')
     print(f'history is - {history}')
@@ -79,7 +82,6 @@ def stream_chat(message: str, history: list, temperature: float, context_window:
 
     print(f"Conversation is -\n{conversation}")
 
-    llm = gemma2()
 
     resp = llm.chat(
         messages = messages,