diff --git a/litellm/proxy/proxy_server.py b/litellm/proxy/proxy_server.py
index 690c24da4..c9ba4c215 100644
--- a/litellm/proxy/proxy_server.py
+++ b/litellm/proxy/proxy_server.py
@@ -1093,8 +1093,11 @@ async def model_info(request: Request):
         data = copy.deepcopy(model["litellm_params"])
         data["messages"] = [{"role": "user", "content": "Hey, how's it going?"}]
         data["max_tokens"] = 10
+        print(f"data going to litellm acompletion: {data}")
         response = await litellm.acompletion(**data)
-        litellm_model_info = litellm.model_cost.get(response["model"], {})
+        response_model = response["model"]
+        print(f"response model: {response_model}; response - {response}")
+        litellm_model_info = litellm.get_model_info(response_model)
         model_info = model.get("model_info", {})
         for k, v in litellm_model_info.items(): 
             if k not in model_info: 
diff --git a/litellm/utils.py b/litellm/utils.py
index bed6d1cda..280a6342f 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -2703,6 +2703,13 @@ def get_model_info(model: str):
         except requests.exceptions.RequestException as e:
             return None
     try:
+        azure_llms = {
+            "gpt-35-turbo": "azure/gpt-3.5-turbo",
+            "gpt-35-turbo-16k": "azure/gpt-3.5-turbo-16k",
+            "gpt-35-turbo-instruct": "azure/gpt-3.5-turbo-instruct"
+        }
+        if model in azure_llms: 
+            model = azure_llms[model]
         if model in litellm.model_cost:
             return litellm.model_cost[model]
         model, custom_llm_provider, _, _ =  get_llm_provider(model=model)