remove caching on stats route

llm_server/routes/stats.py

@@ -31,4 +31,5 @@ class SemaphoreCheckerThread(Thread):
                 self.values.append(0)
             self.prev_semaphore_value = current_semaphore_value
             proompters_1_min = sum(self.values)
+            # print(proompters_1_min)
             time.sleep(1)

llm_server/routes/v1/proxy.py

@@ -14,8 +14,8 @@ 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):
@@ -23,6 +23,8 @@ def get_stats():
     else:
         online = True
 
+    # print('-->', proompters_1_min)
+
     return jsonify({
         'stats': {
             'proompters_now': opts.concurrent_gens - concurrent_semaphore._value,