diff --git a/app/routes/health.py b/app/routes/health.py
index e7eb9dea..187984cd 100644
--- a/app/routes/health.py
+++ b/app/routes/health.py
@@ -14,7 +14,6 @@ def checkhealth():
     result = []
 
     for model in LLMModel:
-        guidance.llms.OpenAI.cache.clear()
         circuit_status = CircuitBreaker.get_status(
             func=GuidanceWrapper(model=model).is_up,
             cache_key=model,
diff --git a/app/services/guidance_wrapper.py b/app/services/guidance_wrapper.py
index 0a47ac6d..0858f111 100644
--- a/app/services/guidance_wrapper.py
+++ b/app/services/guidance_wrapper.py
@@ -48,6 +48,7 @@ def is_up(self) -> bool:
             True if the model is up, False otherwise.
         """
 
+        guidance.llms.OpenAI.cache.clear()
         handlebars = """
         {{#user~}}Say 1{{~/user}}
         {{#assistant~}}