diff --git a/litellm/llms/anthropic.py b/litellm/llms/anthropic.py
index c90b61a11..2ea07215b 100644
--- a/litellm/llms/anthropic.py
+++ b/litellm/llms/anthropic.py
@@ -141,5 +141,5 @@ class AnthropicLLM:
             }
             return model_response
 
-    def embedding():  # logic for parsing in - calling - parsing out model embedding calls
+    def embedding(self):  # logic for parsing in - calling - parsing out model embedding calls
         pass
diff --git a/litellm/llms/huggingface_restapi.py b/litellm/llms/huggingface_restapi.py
index 974a6c049..23ac16bef 100644
--- a/litellm/llms/huggingface_restapi.py
+++ b/litellm/llms/huggingface_restapi.py
@@ -6,6 +6,7 @@ from litellm import logging
 import time
 from typing import Callable
 from litellm.utils import ModelResponse
+from typing import Optional
 
 
 class HuggingfaceError(Exception):
@@ -45,7 +46,7 @@ class HuggingfaceRestAPILLM:
         logger_fn=None,
     ):  # logic for parsing in - calling - parsing out model completion calls
         if custom_api_base:
-            completion_url = custom_api_base
+            completion_url: Optional[str] = custom_api_base
         elif "HF_API_BASE" in os.environ:
             completion_url = os.getenv("HF_API_BASE")
         else:
@@ -136,5 +137,5 @@ class HuggingfaceRestAPILLM:
             return model_response
         pass
 
-    def embedding():  # logic for parsing in - calling - parsing out model embedding calls
+    def embedding(self):  # logic for parsing in - calling - parsing out model embedding calls
         pass
diff --git a/litellm/main.py b/litellm/main.py
index 713a21ed6..1fc89e2e7 100644
--- a/litellm/main.py
+++ b/litellm/main.py
@@ -265,8 +265,8 @@ def completion(
                     or api_key
                     or litellm.replicate_key
                 )
-                # set replicate kye
-                os.environ["REPLICATE_API_TOKEN"] = replicate_key
+                # set replicate key
+                os.environ["REPLICATE_API_TOKEN"]: str = replicate_key
             prompt = " ".join([message["content"] for message in messages])
             input = {"prompt": prompt}
             if "max_tokens" in optional_params:
diff --git a/litellm/testing.py b/litellm/testing.py
index 3e3ce286e..5db01d182 100644
--- a/litellm/testing.py
+++ b/litellm/testing.py
@@ -101,7 +101,7 @@ def duration_test_model(original_function):
 
 
 @duration_test_model
-def load_test_model(models: list, prompt: str = None, num_calls: int = None):
+def load_test_model(models: list, prompt: str = "", num_calls: int = 0):
     test_calls = 100
     if num_calls:
         test_calls = num_calls
diff --git a/litellm/tests/test_api_key_param.py b/litellm/tests/test_api_key_param.py
index cebcb1a37..c444b3904 100644
--- a/litellm/tests/test_api_key_param.py
+++ b/litellm/tests/test_api_key_param.py
@@ -22,7 +22,7 @@ messages = [{"content": user_message, "role": "user"}]
 
 ## Test 1: Setting key dynamically
 temp_key = os.environ.get("ANTHROPIC_API_KEY")
-os.environ["ANTHROPIC_API_KEY"] = "bad-key"
+os.environ["ANTHROPIC_API_KEY"]: str = "bad-key"
 # test on openai completion call
 try:
     response = completion(
@@ -39,7 +39,7 @@ os.environ["ANTHROPIC_API_KEY"] = temp_key
 
 
 ## Test 2: Setting key via __init__ params
-litellm.anthropic_key = os.environ.get("ANTHROPIC_API_KEY")
+litellm.anthropic_key: str = os.environ.get("ANTHROPIC_API_KEY")
 os.environ.pop("ANTHROPIC_API_KEY")
 # test on openai completion call
 try:
diff --git a/litellm/utils.py b/litellm/utils.py
index a14b5f8f1..b79067c36 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -21,7 +21,7 @@ from .exceptions import (
     ServiceUnavailableError,
     OpenAIError,
 )
-from typing import List, Dict, Union
+from typing import List, Dict, Union, Optional
 
 ####### ENVIRONMENT VARIABLES ###################
 dotenv.load_dotenv()  # Loading env variables using dotenv
@@ -35,10 +35,10 @@ heliconeLogger = None
 aispendLogger = None
 berrispendLogger = None
 supabaseClient = None
-callback_list = []
+callback_list: Optional[List[str]] = []
 user_logger_fn = None
-additional_details = {}
-local_cache = {}
+additional_details: Optional[Dict[str, str]] = {}
+local_cache: Optional[Dict[str, str]] = {}
 
 ######## Model Response #########################
 # All liteLLM Model responses will be in this format, Follows the OpenAI Format