diff --git a/poetry.lock b/poetry.lock
index 2f94693e61..cca2f49a75 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -1823,13 +1823,13 @@ signedtoken = ["cryptography (>=3.0.0)", "pyjwt (>=2.0.0,<3)"]
 
 [[package]]
 name = "openai"
-version = "1.54.0"
+version = "1.56.1"
 description = "The official Python library for the openai API"
 optional = false
 python-versions = ">=3.8"
 files = [
-    {file = "openai-1.54.0-py3-none-any.whl", hash = "sha256:24ed8874b56e919f0fbb80b7136c3fb022dc82ce9f5f21579b7b280ea4bba249"},
-    {file = "openai-1.54.0.tar.gz", hash = "sha256:df2a84384314165b706722a7ac8988dc33eba20dd7fc3b939d138110e608b1ce"},
+    {file = "openai-1.56.1-py3-none-any.whl", hash = "sha256:38e61183c2a98fedebbbb04a909a052d9f897358b070483fc0caff17300a227c"},
+    {file = "openai-1.56.1.tar.gz", hash = "sha256:8b0449f22a0c318441eae8a8a789753c3b2cac86542be51ca45df788e26aa180"},
 ]
 
 [package.dependencies]
@@ -3519,4 +3519,4 @@ proxy = ["PyJWT", "apscheduler", "backoff", "cryptography", "fastapi", "fastapi-
 [metadata]
 lock-version = "2.0"
 python-versions = ">=3.8.1,<4.0, !=3.9.7"
-content-hash = "64154f16e1bbea8b77ba3eddf1cbf051af39f019820d92b638c448445fa32c83"
+content-hash = "1ba487e48917ce3087fa154534c6c495dad7dc8dc1c5881bbb1f196ba6702b91"
diff --git a/pyproject.toml b/pyproject.toml
index c84ff303b5..ad605f2190 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -17,7 +17,8 @@ documentation = "https://docs.litellm.ai"
 
 [tool.poetry.dependencies]
 python = ">=3.8.1,<4.0, !=3.9.7"
-openai = ">=1.54.0"
+httpx = ">=0.23.0,<0.28.0"
+openai = ">=1.55.3"
 python-dotenv = ">=0.2.0"
 tiktoken = ">=0.7.0"
 importlib-metadata = ">=6.8.0"
diff --git a/requirements.txt b/requirements.txt
index b22edea09d..7b1abcdc68 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,5 +1,6 @@
 # LITELLM PROXY DEPENDENCIES #
 anyio==4.4.0 # openai + http req.
+httpx==0.27.0 # Pin Httpx dependency
 openai==1.55.3  # openai req. 
 fastapi==0.111.0 # server dep
 backoff==2.2.1 # server dep
diff --git a/tests/llm_translation/base_llm_unit_tests.py b/tests/llm_translation/base_llm_unit_tests.py
index 028a02535f..a4551378eb 100644
--- a/tests/llm_translation/base_llm_unit_tests.py
+++ b/tests/llm_translation/base_llm_unit_tests.py
@@ -63,6 +63,20 @@ class BaseLLMChatTest(ABC):
         response = litellm.completion(**base_completion_call_args, messages=messages)
         assert response is not None
 
+    def test_multilingual_requests(self):
+        """
+        Tests that the provider can handle multilingual requests and invalid utf-8 sequences
+
+        Context: https://github.com/openai/openai-python/issues/1921
+        """
+        base_completion_call_args = self.get_base_completion_call_args()
+        response = litellm.completion(
+            **base_completion_call_args,
+            messages=[{"role": "user", "content": "你好世界！\ud83e, ö"}],
+        )
+        print("multilingual response: ", response)
+        assert response is not None
+
     @pytest.mark.parametrize(
         "response_format",
         [
diff --git a/tests/llm_translation/test_anthropic_completion.py b/tests/llm_translation/test_anthropic_completion.py
index 076219961b..a61bdfd731 100644
--- a/tests/llm_translation/test_anthropic_completion.py
+++ b/tests/llm_translation/test_anthropic_completion.py
@@ -706,6 +706,14 @@ class TestAnthropicCompletion(BaseLLMChatTest):
         result = convert_to_anthropic_tool_invoke([tool_call_no_arguments])
         print(result)
 
+    def test_multilingual_requests(self):
+        """
+        Anthropic API raises a 400 BadRequest error when the request contains invalid utf-8 sequences.
+
+        Todo: if litellm.modify_params is True ensure it's a valid utf-8 sequence
+        """
+        pass
+
 
 def test_convert_tool_response_to_message_with_values():
     """Test converting a tool response with 'values' key to a message"""
diff --git a/tests/llm_translation/test_openai.py b/tests/llm_translation/test_openai.py
index b07f4c5d23..87e3a8a614 100644
--- a/tests/llm_translation/test_openai.py
+++ b/tests/llm_translation/test_openai.py
@@ -15,6 +15,7 @@ from respx import MockRouter
 
 import litellm
 from litellm import Choices, Message, ModelResponse
+from base_llm_unit_tests import BaseLLMChatTest
 
 
 def test_openai_prediction_param():
@@ -268,3 +269,12 @@ async def test_vision_with_custom_model():
         ]
         assert request_body["model"] == "my-custom-model"
         assert request_body["max_tokens"] == 10
+
+
+class TestOpenAIChatCompletion(BaseLLMChatTest):
+    def get_base_completion_call_args(self) -> dict:
+        return {"model": "gpt-4o-mini"}
+
+    def test_tool_call_no_arguments(self, tool_call_no_arguments):
+        """Test that tool calls with no arguments is translated correctly. Relevant issue: https://github.com/BerriAI/litellm/issues/6833"""
+        pass