diff --git a/llm_server/routes/v1/generate_stats.py b/llm_server/routes/v1/generate_stats.py
index 94d663d..17b8bd4 100644
--- a/llm_server/routes/v1/generate_stats.py
+++ b/llm_server/routes/v1/generate_stats.py
@@ -13,7 +13,9 @@ from llm_server.routes.stats import calculate_avg_gen_time, get_active_gen_worke
 
 def calculate_wait_time(gen_time_calc, proompters_in_queue, concurrent_gens, active_gen_workers):
     workers_running = gen_time_calc if active_gen_workers > 0 else 0
-    if proompters_in_queue > 0:
+    if proompters_in_queue < concurrent_gens:
+        return 0
+    elif proompters_in_queue >= concurrent_gens:
         # Calculate how long it will take to complete the currently running gens and the queued requests.
         # If the proompters in the queue are equal to the number of workers, just use the calculated generation time.
         # Otherwise, use how many requests we can process concurrently times the calculated generation time. Then, round