Update app.py

app.py

@@ -12,6 +12,7 @@ if not os.path.exists(OLLAMA):
     subprocess.run("curl -L https://ollama.com/download/ollama-linux-amd64 -o ~/ollama", shell=True)
     os.chmod(OLLAMA, 0o755)
 
+@spaces.GPU()
 def ollama_service_thread():
     subprocess.run("~/ollama serve", shell=True)
 
@@ -59,7 +60,7 @@ h3 {
 """
 
 # Remove this if in CPU
-@spaces.GPU()
+
 def stream_chat(message: str, history: list, temperature: float, max_new_tokens: int, top_p: float, top_k: int, penalty: float):
     
     conversation = []