Fix bedrock passing response_format: {"type": "text"} (#8900)

* fix(converse_transformation.py): ignore type: text, value in response_format

no-op for bedrock

* fix(converse_transformation.py): handle adding response format value to tools

* fix(base_invoke_transformation.py): fix 'get_bedrock_invoke_provider' to handle cross-region-inferencing models

* test(test_bedrock_completion.py): add unit testing for bedrock invoke provider logic

* test: update test

* fix(exception_mapping_utils.py): add context window exceeded error handling for databricks provider route

* fix(fireworks_ai/): support passing tools + response_format together

* fix: cleanup

* fix(base_invoke_transformation.py): fix imports

litellm/litellm_core_utils/exception_mapping_utils.py

@@ -278,6 +278,7 @@ def exception_type(  # type: ignore  # noqa: PLR0915
                     "This model's maximum context length is" in error_str
                     or "string too long. Expected a string with maximum length"
                     in error_str
+                    or "model's maximum context limit" in error_str
                 ):
                     exception_mapping_worked = True
                     raise ContextWindowExceededError(
@@ -692,6 +693,13 @@ def exception_type(  # type: ignore  # noqa: PLR0915
                         response=getattr(original_exception, "response", None),
                         litellm_debug_info=extra_information,
                     )
+                elif "model's maximum context limit" in error_str:
+                    exception_mapping_worked = True
+                    raise ContextWindowExceededError(
+                        message=f"{custom_llm_provider}Exception: Context Window Error - {error_str}",
+                        model=model,
+                        llm_provider=custom_llm_provider,
+                    )
                 elif "token_quota_reached" in error_str:
                     exception_mapping_worked = True
                     raise RateLimitError(

litellm/llms/anthropic/chat/transformation.py

@@ -293,18 +293,6 @@ class AnthropicConfig(BaseConfig):
                 new_stop = new_v
         return new_stop
 
-    def _add_tools_to_optional_params(
-        self, optional_params: dict, tools: List[AllAnthropicToolsValues]
-    ) -> dict:
-        if "tools" not in optional_params:
-            optional_params["tools"] = tools
-        else:
-            optional_params["tools"] = [
-                *optional_params["tools"],
-                *tools,
-            ]
-        return optional_params
-
     def map_openai_params(
         self,
         non_default_params: dict,

litellm/llms/base_llm/chat/transformation.py

@@ -111,6 +111,19 @@ class BaseConfig(ABC):
         """
         return False
 
+    def _add_tools_to_optional_params(self, optional_params: dict, tools: List) -> dict:
+        """
+        Helper util to add tools to optional_params.
+        """
+        if "tools" not in optional_params:
+            optional_params["tools"] = tools
+        else:
+            optional_params["tools"] = [
+                *optional_params["tools"],
+                *tools,
+            ]
+        return optional_params
+
     def translate_developer_role_to_system_role(
         self,
         messages: List[AllMessageValues],
@@ -158,6 +171,7 @@ class BaseConfig(ABC):
         optional_params: dict,
         value: dict,
         is_response_format_supported: bool,
+        enforce_tool_choice: bool = True,
     ) -> dict:
         """
         Follow similar approach to anthropic - translate to a single tool call.
@@ -195,9 +209,11 @@ class BaseConfig(ABC):
 
             optional_params.setdefault("tools", [])
             optional_params["tools"].append(_tool)
+            if enforce_tool_choice:
                 optional_params["tool_choice"] = _tool_choice
+
             optional_params["json_mode"] = True
-        else:
+        elif is_response_format_supported:
             optional_params["response_format"] = value
         return optional_params
 

litellm/llms/bedrock/chat/converse_transformation.py

@@ -227,6 +227,10 @@ class AmazonConverseConfig(BaseConfig):
                     json_schema = value["json_schema"]["schema"]
                     schema_name = value["json_schema"]["name"]
                     description = value["json_schema"].get("description")
+
+                if "type" in value and value["type"] == "text":
+                    continue
+
                 """
                 Follow similar approach to anthropic - translate to a single tool call. 
 
@@ -240,7 +244,9 @@ class AmazonConverseConfig(BaseConfig):
                     schema_name=schema_name if schema_name != "" else "json_tool_call",
                     description=description,
                 )
-                optional_params["tools"] = [_tool]
+                optional_params = self._add_tools_to_optional_params(
+                    optional_params=optional_params, tools=[_tool]
+                )
                 if litellm.utils.supports_tool_choice(
                     model=model, custom_llm_provider=self.custom_llm_provider
                 ):
@@ -267,7 +273,9 @@ class AmazonConverseConfig(BaseConfig):
             if param == "top_p":
                 optional_params["topP"] = value
             if param == "tools":
-                optional_params["tools"] = value
+                optional_params = self._add_tools_to_optional_params(
+                    optional_params=optional_params, tools=value
+                )
             if param == "tool_choice":
                 _tool_choice_value = self.map_tool_choice_values(
                     model=model, tool_choice=value, drop_params=drop_params  # type: ignore

litellm/llms/bedrock/chat/invoke_transformations/base_invoke_transformation.py

@@ -3,7 +3,7 @@ import json
 import time
 import urllib.parse
 from functools import partial
-from typing import TYPE_CHECKING, Any, List, Optional, Tuple, Union
+from typing import TYPE_CHECKING, Any, List, Optional, Tuple, Union, cast, get_args
 
 import httpx
 
@@ -531,6 +531,60 @@ class AmazonInvokeConfig(BaseConfig, BaseAWSLLM):
         """
         return False
 
+    @staticmethod
+    def get_bedrock_invoke_provider(
+        model: str,
+    ) -> Optional[litellm.BEDROCK_INVOKE_PROVIDERS_LITERAL]:
+        """
+        Helper function to get the bedrock provider from the model
+
+        handles 4 scenarios:
+        1. model=invoke/anthropic.claude-3-5-sonnet-20240620-v1:0 -> Returns `anthropic`
+        2. model=anthropic.claude-3-5-sonnet-20240620-v1:0 -> Returns `anthropic`
+        3. model=llama/arn:aws:bedrock:us-east-1:086734376398:imported-model/r4c4kewx2s0n -> Returns `llama`
+        4. model=us.amazon.nova-pro-v1:0 -> Returns `nova`
+        """
+        if model.startswith("invoke/"):
+            model = model.replace("invoke/", "", 1)
+
+        _split_model = model.split(".")[0]
+        if _split_model in get_args(litellm.BEDROCK_INVOKE_PROVIDERS_LITERAL):
+            return cast(litellm.BEDROCK_INVOKE_PROVIDERS_LITERAL, _split_model)
+
+        # If not a known provider, check for pattern with two slashes
+        provider = AmazonInvokeConfig._get_provider_from_model_path(model)
+        if provider is not None:
+            return provider
+
+        # check if provider == "nova"
+        if "nova" in model:
+            return "nova"
+
+        for provider in get_args(litellm.BEDROCK_INVOKE_PROVIDERS_LITERAL):
+            if provider in model:
+                return provider
+        return None
+
+    @staticmethod
+    def _get_provider_from_model_path(
+        model_path: str,
+    ) -> Optional[litellm.BEDROCK_INVOKE_PROVIDERS_LITERAL]:
+        """
+        Helper function to get the provider from a model path with format: provider/model-name
+
+        Args:
+            model_path (str): The model path (e.g., 'llama/arn:aws:bedrock:us-east-1:086734376398:imported-model/r4c4kewx2s0n' or 'anthropic/model-name')
+
+        Returns:
+            Optional[str]: The provider name, or None if no valid provider found
+        """
+        parts = model_path.split("/")
+        if len(parts) >= 1:
+            provider = parts[0]
+            if provider in get_args(litellm.BEDROCK_INVOKE_PROVIDERS_LITERAL):
+                return cast(litellm.BEDROCK_INVOKE_PROVIDERS_LITERAL, provider)
+        return None
+
     def get_bedrock_model_id(
         self,
         optional_params: dict,

litellm/llms/fireworks_ai/chat/transformation.py

@@ -90,6 +90,11 @@ class FireworksAIConfig(OpenAIGPTConfig):
     ) -> dict:
 
         supported_openai_params = self.get_supported_openai_params(model=model)
+        is_tools_set = any(
+            param == "tools" and value is not None
+            for param, value in non_default_params.items()
+        )
+
         for param, value in non_default_params.items():
             if param == "tool_choice":
                 if value == "required":
@@ -98,18 +103,30 @@ class FireworksAIConfig(OpenAIGPTConfig):
                 else:
                     # pass through the value of tool choice
                     optional_params["tool_choice"] = value
-            elif (
-                param == "response_format" and value.get("type", None) == "json_schema"
-            ):
+            elif param == "response_format":
+
+                if (
+                    is_tools_set
+                ):  # fireworks ai doesn't support tools and response_format together
+                    optional_params = self._add_response_format_to_tools(
+                        optional_params=optional_params,
+                        value=value,
+                        is_response_format_supported=False,
+                        enforce_tool_choice=False,  # tools and response_format are both set, don't enforce tool_choice
+                    )
+                elif "json_schema" in value:
                     optional_params["response_format"] = {
                         "type": "json_object",
                         "schema": value["json_schema"]["schema"],
                     }
+                else:
+                    optional_params["response_format"] = value
             elif param == "max_completion_tokens":
                 optional_params["max_tokens"] = value
             elif param in supported_openai_params:
                 if value is not None:
                     optional_params[param] = value
+
         return optional_params
 
     def _add_transform_inline_image_block(

tests/llm_translation/base_llm_unit_tests.py

@@ -254,6 +254,56 @@ class BaseLLMChatTest(ABC):
         # relevant issue: https://github.com/BerriAI/litellm/issues/6741
         assert response.choices[0].message.content is not None
 
+
+    @pytest.mark.parametrize(
+        "response_format",
+        [
+            {"type": "text"},
+        ],
+    )
+    @pytest.mark.flaky(retries=6, delay=1)
+    def test_response_format_type_text_with_tool_calls_no_tool_choice(
+        self, response_format
+    ):
+        base_completion_call_args = self.get_base_completion_call_args()
+        messages = [
+            {"role": "user", "content": "What's the weather like in Boston today?"},
+        ]
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_current_weather",
+                    "description": "Get the current weather in a given location",
+                    "parameters": {
+                        "type": "object",
+                        "properties": {
+                            "location": {
+                                "type": "string",
+                                "description": "The city and state, e.g. San Francisco, CA",
+                            },
+                            "unit": {
+                                "type": "string",
+                                "enum": ["celsius", "fahrenheit"],
+                            },
+                        },
+                        "required": ["location"],
+                    },
+                },
+            }
+        ]
+        try:
+            response = self.completion_function(
+                **base_completion_call_args,
+                messages=messages,
+                response_format=response_format,
+                tools=tools,
+                drop_params=True,
+            )
+        except litellm.ContextWindowExceededError:
+            pytest.skip("Model exceeded context window")
+        assert response is not None
+
     def test_response_format_type_text(self):
         """
         Test that the response format type text does not lead to tool calls
@@ -287,6 +337,7 @@ class BaseLLMChatTest(ABC):
 
         print(f"translated_params={translated_params}")
 
+
     @pytest.mark.flaky(retries=6, delay=1)
     def test_json_response_pydantic_obj(self):
         litellm.set_verbose = True

tests/llm_translation/test_bedrock_completion.py

@@ -2717,6 +2717,33 @@ def test_bedrock_top_k_param(model, expected_params):
             assert data["additionalModelRequestFields"] == expected_params
 
 
+
+def test_bedrock_invoke_provider():
+    assert (
+        litellm.AmazonInvokeConfig().get_bedrock_invoke_provider(
+            "bedrock/invoke/us.anthropic.claude-3-5-sonnet-20240620-v1:0"
+        )
+        == "anthropic"
+    )
+    assert (
+        litellm.AmazonInvokeConfig().get_bedrock_invoke_provider(
+            "bedrock/us.anthropic.claude-3-5-sonnet-20240620-v1:0"
+        )
+        == "anthropic"
+    )
+    assert (
+        litellm.AmazonInvokeConfig().get_bedrock_invoke_provider(
+            "bedrock/llama/arn:aws:bedrock:us-east-1:086734376398:imported-model/r4c4kewx2s0n"
+        )
+        == "llama"
+    )
+    assert (
+        litellm.AmazonInvokeConfig().get_bedrock_invoke_provider(
+            "us.amazon.nova-pro-v1:0"
+        )
+        == "nova"
+    )
+
 def test_bedrock_description_param():
     from litellm import completion
     from litellm.llms.custom_httpx.http_handler import HTTPHandler
@@ -2754,3 +2781,4 @@ def test_bedrock_description_param():
         assert (
             "Find the meaning inside a poem" in request_body_str
         )  # assert description is passed
+