fix(bedrock.py): convert httpx.timeout to boto3 valid timeout

Closes https://github.com/BerriAI/litellm/issues/3398

litellm/llms/azure.py

@@ -151,7 +151,7 @@ class AzureChatCompletion(BaseLLM):
         api_type: str,
         azure_ad_token: str,
         print_verbose: Callable,
-        timeout,
+        timeout: Union[float, httpx.Timeout],
         logging_obj,
         optional_params,
         litellm_params,

litellm/llms/bedrock.py

@@ -533,7 +533,7 @@ def init_bedrock_client(
     aws_session_name: Optional[str] = None,
     aws_profile_name: Optional[str] = None,
     aws_role_name: Optional[str] = None,
-    timeout: Optional[int] = None,
+    timeout: Optional[Union[float, httpx.Timeout]] = None,
 ):
     # check for custom AWS_REGION_NAME and use it if not passed to init_bedrock_client
     litellm_aws_region_name = get_secret("AWS_REGION_NAME", None)
@@ -592,7 +592,12 @@ def init_bedrock_client(
 
     import boto3
 
-    config = boto3.session.Config(connect_timeout=timeout, read_timeout=timeout)
+    if isinstance(timeout, float):
+        config = boto3.session.Config(connect_timeout=timeout, read_timeout=timeout)
+    elif isinstance(timeout, httpx.Timeout):
+        config = boto3.session.Config(
+            connect_timeout=timeout.connect, read_timeout=timeout.read
+        )
 
     ### CHECK STS ###
     if aws_role_name is not None and aws_session_name is not None:

litellm/llms/openai.py

@@ -246,7 +246,7 @@ class OpenAIChatCompletion(BaseLLM):
     def completion(
         self,
         model_response: ModelResponse,
-        timeout: float,
+        timeout: Union[float, httpx.Timeout],
         model: Optional[str] = None,
         messages: Optional[list] = None,
         print_verbose: Optional[Callable] = None,
@@ -271,9 +271,12 @@ class OpenAIChatCompletion(BaseLLM):
             if model is None or messages is None:
                 raise OpenAIError(status_code=422, message=f"Missing model or messages")
 
-            if not isinstance(timeout, float):
+            if not isinstance(timeout, float) and not isinstance(
+                timeout, httpx.Timeout
+            ):
                 raise OpenAIError(
-                    status_code=422, message=f"Timeout needs to be a float"
+                    status_code=422,
+                    message=f"Timeout needs to be a float or httpx.Timeout",
                 )
 
             if custom_llm_provider != "openai":
@@ -425,7 +428,7 @@ class OpenAIChatCompletion(BaseLLM):
         self,
         data: dict,
         model_response: ModelResponse,
-        timeout: float,
+        timeout: Union[float, httpx.Timeout],
         api_key: Optional[str] = None,
         api_base: Optional[str] = None,
         organization: Optional[str] = None,
@@ -480,7 +483,7 @@ class OpenAIChatCompletion(BaseLLM):
     def streaming(
         self,
         logging_obj,
-        timeout: float,
+        timeout: Union[float, httpx.Timeout],
         data: dict,
         model: str,
         api_key: Optional[str] = None,
@@ -524,7 +527,7 @@ class OpenAIChatCompletion(BaseLLM):
     async def async_streaming(
         self,
         logging_obj,
-        timeout: float,
+        timeout: Union[float, httpx.Timeout],
         data: dict,
         model: str,
         api_key: Optional[str] = None,

litellm/main.py

@@ -39,6 +39,7 @@ from litellm.utils import (
     Usage,
     get_optional_params_embeddings,
     get_optional_params_image_gen,
+    supports_httpx_timeout,
 )
 from .llms import (
     anthropic_text,
@@ -450,7 +451,7 @@ def completion(
     model: str,
     # Optional OpenAI params: see https://platform.openai.com/docs/api-reference/chat/create
     messages: List = [],
-    timeout: Optional[Union[float, int]] = None,
+    timeout: Optional[Union[float, str, httpx.Timeout]] = None,
     temperature: Optional[float] = None,
     top_p: Optional[float] = None,
     n: Optional[int] = None,
@@ -648,11 +649,21 @@ def completion(
     non_default_params = {
         k: v for k, v in kwargs.items() if k not in default_params
     }  # model-specific params - pass them straight to the model/provider
-    if timeout is None:
-        timeout = (
-            kwargs.get("request_timeout", None) or 600
-        )  # set timeout for 10 minutes by default
-    timeout = float(timeout)
+
+    ### TIMEOUT LOGIC ###
+    timeout = timeout or kwargs.get("request_timeout", 600) or 600
+    # set timeout for 10 minutes by default
+
+    if (
+        timeout is not None
+        and isinstance(timeout, httpx.Timeout)
+        and supports_httpx_timeout(custom_llm_provider) == False
+    ):
+        read_timeout = timeout.read or 600
+        timeout = read_timeout  # default 10 min timeout
+    elif timeout is not None and not isinstance(timeout, httpx.Timeout):
+        timeout = float(timeout)  # type: ignore
+
     try:
         if base_url is not None:
             api_base = base_url
@@ -873,7 +884,7 @@ def completion(
                 logger_fn=logger_fn,
                 logging_obj=logging,
                 acompletion=acompletion,
-                timeout=timeout,
+                timeout=timeout,  # type: ignore
                 client=client,  # pass AsyncAzureOpenAI, AzureOpenAI client
             )
 
@@ -1014,7 +1025,7 @@ def completion(
                     optional_params=optional_params,
                     litellm_params=litellm_params,
                     logger_fn=logger_fn,
-                    timeout=timeout,
+                    timeout=timeout,  # type: ignore
                     custom_prompt_dict=custom_prompt_dict,
                     client=client,  # pass AsyncOpenAI, OpenAI client
                     organization=organization,
@@ -1099,7 +1110,7 @@ def completion(
                 optional_params=optional_params,
                 litellm_params=litellm_params,
                 logger_fn=logger_fn,
-                timeout=timeout,
+                timeout=timeout,  # type: ignore
             )
 
             if (
@@ -1473,7 +1484,7 @@ def completion(
                 acompletion=acompletion,
                 logging_obj=logging,
                 custom_prompt_dict=custom_prompt_dict,
-                timeout=timeout,
+                timeout=timeout,  # type: ignore
             )
             if (
                 "stream" in optional_params
@@ -1566,7 +1577,7 @@ def completion(
                 logger_fn=logger_fn,
                 logging_obj=logging,
                 acompletion=acompletion,
-                timeout=timeout,
+                timeout=timeout,  # type: ignore
             )
             ## LOGGING
             logging.post_call(
@@ -1893,7 +1904,7 @@ def completion(
                 logger_fn=logger_fn,
                 encoding=encoding,
                 logging_obj=logging,
-                timeout=timeout,
+                timeout=timeout,  # type: ignore
             )
             if (
                 "stream" in optional_params

litellm/router.py

@@ -375,7 +375,9 @@ class Router:
         except Exception as e:
             raise e
 
-    def _completion(self, model: str, messages: List[Dict[str, str]], **kwargs):
+    def _completion(
+        self, model: str, messages: List[Dict[str, str]], **kwargs
+    ) -> Union[ModelResponse, CustomStreamWrapper]:
         model_name = None
         try:
             # pick the one that is available (lowest TPM/RPM)
@@ -438,7 +440,9 @@ class Router:
             )
             raise e
 
-    async def acompletion(self, model: str, messages: List[Dict[str, str]], **kwargs):
+    async def acompletion(
+        self, model: str, messages: List[Dict[str, str]], **kwargs
+    ) -> Union[ModelResponse, CustomStreamWrapper]:
         try:
             kwargs["model"] = model
             kwargs["messages"] = messages
@@ -454,7 +458,9 @@ class Router:
         except Exception as e:
             raise e
 
-    async def _acompletion(self, model: str, messages: List[Dict[str, str]], **kwargs):
+    async def _acompletion(
+        self, model: str, messages: List[Dict[str, str]], **kwargs
+    ) -> Union[ModelResponse, CustomStreamWrapper]:
         """
         - Get an available deployment
         - call it with a semaphore over the call

litellm/tests/test_timeout.py

@@ -10,7 +10,37 @@ sys.path.insert(
 import time
 import litellm
 import openai
-import pytest, uuid
+import pytest, uuid, httpx
+
+
+@pytest.mark.parametrize(
+    "model, provider",
+    [
+        ("gpt-3.5-turbo", "openai"),
+        ("anthropic.claude-instant-v1", "bedrock"),
+        ("azure/chatgpt-v-2", "azure"),
+    ],
+)
+@pytest.mark.parametrize("sync_mode", [True, False])
+@pytest.mark.asyncio
+async def test_httpx_timeout(model, provider, sync_mode):
+    """
+    Test if setting httpx.timeout works for completion calls
+    """
+    timeout_val = httpx.Timeout(10.0, connect=60.0)
+
+    messages = [{"role": "user", "content": "Hey, how's it going?"}]
+
+    if sync_mode:
+        response = litellm.completion(
+            model=model, messages=messages, timeout=timeout_val
+        )
+    else:
+        response = await litellm.acompletion(
+            model=model, messages=messages, timeout=timeout_val
+        )
+
+    print(f"response: {response}")
 
 
 def test_timeout():

litellm/utils.py

@@ -4442,7 +4442,19 @@ def completion_cost(
         raise e
 
 
-def supports_function_calling(model: str):
+def supports_httpx_timeout(custom_llm_provider: str) -> bool:
+    """
+    Helper function to know if a provider implementation supports httpx timeout
+    """
+    supported_providers = ["openai", "azure", "bedrock"]
+
+    if custom_llm_provider in supported_providers:
+        return True
+
+    return False
+
+
+def supports_function_calling(model: str) -> bool:
     """
     Check if the given model supports function calling and return a boolean value.