diff --git a/litellm/__init__.py b/litellm/__init__.py
index 0080e9551c..4e1410a319 100644
--- a/litellm/__init__.py
+++ b/litellm/__init__.py
@@ -26,7 +26,6 @@ from litellm._logging import (
     log_level,
 )
 import re
-from .messages import *
 from litellm.constants import (
     DEFAULT_BATCH_SIZE,
     DEFAULT_FLUSH_INTERVAL_SECONDS,
@@ -1028,6 +1027,7 @@ from .proxy.proxy_cli import run_server
 from .router import Router
 from .assistants.main import *
 from .batches.main import *
+from .messages import *
 from .batch_completion.main import *  # type: ignore
 from .rerank_api.main import *
 from .llms.anthropic.experimental_pass_through.messages.handler import *
diff --git a/litellm/llms/anthropic/experimental_pass_through/messages/handler.py b/litellm/llms/anthropic/experimental_pass_through/messages/handler.py
index 0b809a4460..fd7a3d60a8 100644
--- a/litellm/llms/anthropic/experimental_pass_through/messages/handler.py
+++ b/litellm/llms/anthropic/experimental_pass_through/messages/handler.py
@@ -71,7 +71,7 @@ async def anthropic_messages(
     stop_sequences: Optional[List[str]] = None,
     stream: Optional[bool] = False,
     system: Optional[str] = None,
-    temperature: Optional[float] = 1.0,
+    temperature: Optional[float] = None,
     thinking: Optional[Dict] = None,
     tool_choice: Optional[Dict] = None,
     tools: Optional[List[Dict]] = None,
@@ -152,6 +152,7 @@ async def anthropic_messages(
         in anthropic_messages_provider_config.get_supported_anthropic_messages_params(
             model=model
         )
+        and v is not None
     }
     request_body["stream"] = stream
     request_body["model"] = model
diff --git a/litellm/messages/__init__.py b/litellm/messages/__init__.py
index c5cb3f5330..f3249f981b 100644
--- a/litellm/messages/__init__.py
+++ b/litellm/messages/__init__.py
@@ -2,11 +2,17 @@
 Interface for Anthropic's messages API
 
 Use this to call LLMs in Anthropic /messages Request/Response format
+
+This is an __init__.py file to allow the following interface
+
+- litellm.messages.acreate
+- litellm.messages.create
+
 """
 
-from typing import Dict, List, Optional, Union
+from typing import AsyncIterator, Dict, Iterator, List, Optional, Union
 
-from litellm.llms.anthropic.experimental_pass_through.handler import (
+from litellm.llms.anthropic.experimental_pass_through.messages.handler import (
     anthropic_messages as _async_anthropic_messages,
 )
 from litellm.types.llms.anthropic_messages.anthropic_response import (
@@ -29,7 +35,7 @@ async def acreate(
     top_k: Optional[int] = None,
     top_p: Optional[float] = None,
     **kwargs
-) -> AnthropicMessagesResponse:
+) -> Union[AnthropicMessagesResponse, AsyncIterator]:
     """
     Async wrapper for Anthropic's messages API
 
@@ -85,7 +91,7 @@ async def create(
     top_k: Optional[int] = None,
     top_p: Optional[float] = None,
     **kwargs
-) -> AnthropicMessagesResponse:
+) -> Union[AnthropicMessagesResponse, Iterator]:
     """
     Async wrapper for Anthropic's messages API
 
diff --git a/tests/pass_through_unit_tests/test_anthropic_messages_passthrough.py b/tests/pass_through_unit_tests/test_anthropic_messages_passthrough.py
index b5b3302acc..cb7ac0d3bf 100644
--- a/tests/pass_through_unit_tests/test_anthropic_messages_passthrough.py
+++ b/tests/pass_through_unit_tests/test_anthropic_messages_passthrough.py
@@ -8,7 +8,7 @@ import unittest.mock
 from unittest.mock import AsyncMock, MagicMock
 
 sys.path.insert(
-    0, os.path.abspath("../..")
+    0, os.path.abspath("../../..")
 )  # Adds the parent directory to the system path
 import litellm
 import pytest
@@ -16,6 +16,7 @@ from dotenv import load_dotenv
 from litellm.llms.anthropic.experimental_pass_through.messages.handler import (
     anthropic_messages,
 )
+
 from typing import Optional
 from litellm.types.utils import StandardLoggingPayload
 from litellm.integrations.custom_logger import CustomLogger
@@ -73,6 +74,7 @@ async def test_anthropic_messages_non_streaming():
     """
     Test the anthropic_messages with non-streaming request
     """
+    litellm._turn_on_debug()
     # Get API key from environment
     api_key = os.getenv("ANTHROPIC_API_KEY")
     if not api_key:
@@ -82,7 +84,7 @@ async def test_anthropic_messages_non_streaming():
     messages = [{"role": "user", "content": "Hello, can you tell me a short joke?"}]
 
     # Call the handler
-    response = await anthropic_messages(
+    response = await litellm.messages.acreate(
         messages=messages,
         api_key=api_key,
         model="claude-3-haiku-20240307",
@@ -114,7 +116,7 @@ async def test_anthropic_messages_streaming():
 
     # Call the handler
     async_httpx_client = AsyncHTTPHandler()
-    response = await anthropic_messages(
+    response = await litellm.messages.acreate(
         messages=messages,
         api_key=api_key,
         model="claude-3-haiku-20240307",
@@ -134,7 +136,7 @@ async def test_anthropic_messages_streaming_with_bad_request():
     Test the anthropic_messages with streaming request
     """
     try:
-        response = await anthropic_messages(
+        response = await litellm.messages.acreate(
             messages=["hi"],
             api_key=os.getenv("ANTHROPIC_API_KEY"),
             model="claude-3-haiku-20240307",
@@ -458,7 +460,7 @@ async def test_anthropic_messages_with_extra_headers():
     mock_client.post = AsyncMock(return_value=mock_response)
 
     # Call the handler with extra_headers and our mocked client
-    response = await anthropic_messages(
+    response = await litellm.messages.acreate(
         messages=messages,
         api_key=api_key,
         model="claude-3-haiku-20240307",