diff --git a/server/__init__.py b/model_server/__init__.py
similarity index 100%
rename from server/__init__.py
rename to model_server/__init__.py
diff --git a/server/server.py b/model_server/server.py
similarity index 100%
rename from server/server.py
rename to model_server/server.py
diff --git a/scripts/run_server.py b/scripts/run_server.py
index 602a631..d3b748c 100644
--- a/scripts/run_server.py
+++ b/scripts/run_server.py
@@ -1,4 +1,4 @@
-from server import ModelServer
+from model_server import ModelServer
 import argparse
 
 if __name__ == '__main__':
diff --git a/text-generation-webui/custom_model_server_monkey_patch.py b/text-generation-webui/custom_model_server_monkey_patch.py
index ea76f8f..840e16f 100644
--- a/text-generation-webui/custom_model_server_monkey_patch.py
+++ b/text-generation-webui/custom_model_server_monkey_patch.py
@@ -1,4 +1,4 @@
-from server import ModelClient
+from model_server import ModelClient
 from transformers import LlamaTokenizer
 
 def load_model_llama(*args, **kwargs):
diff --git a/text-generation-webui/generate_monkey_patch.py b/text-generation-webui/generate_monkey_patch.py
index 3848840..1d5dc85 100644
--- a/text-generation-webui/generate_monkey_patch.py
+++ b/text-generation-webui/generate_monkey_patch.py
@@ -1,6 +1,6 @@
 import modules.text_generation
 from modules.text_generation import *
-from server import _SentinelTokenStoppingCriteria
+from model_server import _SentinelTokenStoppingCriteria
 
 def generate_reply_patched(question, state, eos_token=None, stopping_strings=[]):
     if shared.model_name == 'None' or shared.model is None: