cache again

llm_server/routes/v1/proxy.py

@@ -7,13 +7,15 @@ from llm_server import opts
 from llm_server.routes.v1.generate import concurrent_semaphore
 from . import bp
 from .. import stats
+from ..cache import cache
+from ..helpers.http import cache_control
 from ..stats import SemaphoreCheckerThread
 from ...llm.info import get_running_model
 
 
 @bp.route('/stats', methods=['GET'])
-# @cache.cached(timeout=5, query_string=True)
-# @cache_control(5)
+@cache.cached(timeout=5, query_string=True)
+@cache_control(5)
 def get_stats():
     model_list = get_running_model()  # will return False when the fetch fails
     if isinstance(model_list, bool):