diff --git a/llm_server/llm/openai/oai_to_vllm.py b/llm_server/llm/openai/oai_to_vllm.py
index 9111389..e224418 100644
--- a/llm_server/llm/openai/oai_to_vllm.py
+++ b/llm_server/llm/openai/oai_to_vllm.py
@@ -26,6 +26,7 @@ def oai_to_vllm(request_json_body, hashes: bool, mode):
 
 
 def format_oai_err(err_msg):
+    print('OAI ERROR MESSAGE:', err_msg)
     return jsonify({
         "error": {
             "message": err_msg,
diff --git a/llm_server/routes/openai_request_handler.py b/llm_server/routes/openai_request_handler.py
index 8664695..0dfd558 100644
--- a/llm_server/routes/openai_request_handler.py
+++ b/llm_server/routes/openai_request_handler.py
@@ -25,6 +25,7 @@ class OpenAIRequestHandler(RequestHandler):
         self.prompt = None
 
     def handle_request(self) -> Tuple[flask.Response, int]:
+        print('recieved request')
         assert not self.used
 
         if opts.openai_silent_trim:
@@ -66,11 +67,13 @@ class OpenAIRequestHandler(RequestHandler):
         model = self.request_json_body.get('model')
 
         if success:
+            print('sent success response')
             return self.build_openai_response(self.prompt, backend_response.json['results'][0]['text'], model=model), backend_response_status_code
         else:
             return backend_response, backend_response_status_code
 
     def handle_ratelimited(self, do_log: bool = True):
+        print('OAI ratelimited:', self.client_ip, self.request.headers)
         _, default_backend_info = get_model_choices()
         w = int(default_backend_info['estimated_wait']) if default_backend_info['estimated_wait'] > 0 else 2
         response = jsonify({