diff --git a/litellm/litellm_core_utils/get_supported_openai_params.py b/litellm/litellm_core_utils/get_supported_openai_params.py
index a832605b8e..ca08df1858 100644
--- a/litellm/litellm_core_utils/get_supported_openai_params.py
+++ b/litellm/litellm_core_utils/get_supported_openai_params.py
@@ -1,8 +1,9 @@
 from typing import Literal, Optional
 
 import litellm
-from litellm import LlmProviders
+from litellm._logging import verbose_logger
 from litellm.exceptions import BadRequestError
+from litellm.types.utils import LlmProviders, LlmProvidersSet
 
 
 def get_supported_openai_params(  # noqa: PLR0915
@@ -30,6 +31,23 @@ def get_supported_openai_params(  # noqa: PLR0915
         except BadRequestError:
             return None
 
+    if custom_llm_provider in LlmProvidersSet:
+        provider_config = litellm.ProviderConfigManager.get_provider_chat_config(
+            model=model, provider=LlmProviders(custom_llm_provider)
+        )
+    elif custom_llm_provider.split("/")[0] in LlmProvidersSet:
+        provider_config = litellm.ProviderConfigManager.get_provider_chat_config(
+            model=model, provider=LlmProviders(custom_llm_provider.split("/")[0])
+        )
+    else:
+        provider_config = None
+
+    if provider_config and request_type == "chat_completion":
+        verbose_logger.info(
+            f"using provider_config: {provider_config} for checking supported params"
+        )
+        return provider_config.get_supported_openai_params(model=model)
+
     if custom_llm_provider == "bedrock":
         return litellm.AmazonConverseConfig().get_supported_openai_params(model=model)
     elif custom_llm_provider == "ollama":
@@ -226,7 +244,8 @@ def get_supported_openai_params(  # noqa: PLR0915
             provider_config = litellm.ProviderConfigManager.get_provider_chat_config(
                 model=model, provider=LlmProviders.CUSTOM
             )
-            return provider_config.get_supported_openai_params(model=model)
+            if provider_config:
+                return provider_config.get_supported_openai_params(model=model)
         elif request_type == "embeddings":
             return None
         elif request_type == "transcription":
diff --git a/litellm/litellm_core_utils/prompt_templates/common_utils.py b/litellm/litellm_core_utils/prompt_templates/common_utils.py
index 8d3845969a..0f2d0da388 100644
--- a/litellm/litellm_core_utils/prompt_templates/common_utils.py
+++ b/litellm/litellm_core_utils/prompt_templates/common_utils.py
@@ -348,3 +348,36 @@ def update_messages_with_model_file_ids(
                             )
                             file_object_file_field["file_id"] = provider_file_id
     return messages
+
+
+def unpack_defs(schema, defs):
+    properties = schema.get("properties", None)
+    if properties is None:
+        return
+
+    for name, value in properties.items():
+        ref_key = value.get("$ref", None)
+        if ref_key is not None:
+            ref = defs[ref_key.split("defs/")[-1]]
+            unpack_defs(ref, defs)
+            properties[name] = ref
+            continue
+
+        anyof = value.get("anyOf", None)
+        if anyof is not None:
+            for i, atype in enumerate(anyof):
+                ref_key = atype.get("$ref", None)
+                if ref_key is not None:
+                    ref = defs[ref_key.split("defs/")[-1]]
+                    unpack_defs(ref, defs)
+                    anyof[i] = ref
+            continue
+
+        items = value.get("items", None)
+        if items is not None:
+            ref_key = items.get("$ref", None)
+            if ref_key is not None:
+                ref = defs[ref_key.split("defs/")[-1]]
+                unpack_defs(ref, defs)
+                value["items"] = ref
+                continue
diff --git a/litellm/litellm_core_utils/prompt_templates/factory.py b/litellm/litellm_core_utils/prompt_templates/factory.py
index e8d8456ed7..1495c05685 100644
--- a/litellm/litellm_core_utils/prompt_templates/factory.py
+++ b/litellm/litellm_core_utils/prompt_templates/factory.py
@@ -3442,6 +3442,8 @@ def _bedrock_tools_pt(tools: List) -> List[BedrockToolBlock]:
         }
     ]
     """
+    from litellm.litellm_core_utils.prompt_templates.common_utils import unpack_defs
+
     tool_block_list: List[BedrockToolBlock] = []
     for tool in tools:
         parameters = tool.get("function", {}).get(
@@ -3455,6 +3457,13 @@ def _bedrock_tools_pt(tools: List) -> List[BedrockToolBlock]:
         description = tool.get("function", {}).get(
             "description", name
         )  # converse api requires a description
+
+        defs = parameters.pop("$defs", {})
+        defs_copy = copy.deepcopy(defs)
+        # flatten the defs
+        for _, value in defs_copy.items():
+            unpack_defs(value, defs_copy)
+        unpack_defs(parameters, defs_copy)
         tool_input_schema = BedrockToolInputSchemaBlock(json=parameters)
         tool_spec = BedrockToolSpecBlock(
             inputSchema=tool_input_schema, name=name, description=description
diff --git a/litellm/llms/anthropic/chat/handler.py b/litellm/llms/anthropic/chat/handler.py
index c29a98b217..f2a5542dcd 100644
--- a/litellm/llms/anthropic/chat/handler.py
+++ b/litellm/llms/anthropic/chat/handler.py
@@ -307,6 +307,10 @@ class AnthropicChatCompletion(BaseLLM):
             model=model,
             provider=LlmProviders(custom_llm_provider),
         )
+        if config is None:
+            raise ValueError(
+                f"Provider config not found for model: {model} and provider: {custom_llm_provider}"
+            )
 
         data = config.transform_request(
             model=model,
diff --git a/litellm/llms/bedrock/chat/converse_transformation.py b/litellm/llms/bedrock/chat/converse_transformation.py
index cf0ec1100e..8ce2c4818b 100644
--- a/litellm/llms/bedrock/chat/converse_transformation.py
+++ b/litellm/llms/bedrock/chat/converse_transformation.py
@@ -235,7 +235,6 @@ class AmazonConverseConfig(BaseConfig):
         optional_params: dict,
         model: str,
         drop_params: bool,
-        messages: Optional[List[AllMessageValues]] = None,
     ) -> dict:
         is_thinking_enabled = self.is_thinking_enabled(non_default_params)
 
diff --git a/litellm/llms/bedrock/chat/invoke_transformations/amazon_cohere_transformation.py b/litellm/llms/bedrock/chat/invoke_transformations/amazon_cohere_transformation.py
index e9479c8f32..9c2c95e6ce 100644
--- a/litellm/llms/bedrock/chat/invoke_transformations/amazon_cohere_transformation.py
+++ b/litellm/llms/bedrock/chat/invoke_transformations/amazon_cohere_transformation.py
@@ -1,13 +1,13 @@
 import types
 from typing import List, Optional
 
-from litellm.llms.base_llm.chat.transformation import BaseConfig
 from litellm.llms.bedrock.chat.invoke_transformations.base_invoke_transformation import (
     AmazonInvokeConfig,
 )
+from litellm.llms.cohere.chat.transformation import CohereChatConfig
 
 
-class AmazonCohereConfig(AmazonInvokeConfig, BaseConfig):
+class AmazonCohereConfig(AmazonInvokeConfig, CohereChatConfig):
     """
     Reference: https://us-west-2.console.aws.amazon.com/bedrock/home?region=us-west-2#/providers?model=command
 
@@ -19,7 +19,6 @@ class AmazonCohereConfig(AmazonInvokeConfig, BaseConfig):
     """
 
     max_tokens: Optional[int] = None
-    temperature: Optional[float] = None
     return_likelihood: Optional[str] = None
 
     def __init__(
@@ -55,11 +54,10 @@ class AmazonCohereConfig(AmazonInvokeConfig, BaseConfig):
         }
 
     def get_supported_openai_params(self, model: str) -> List[str]:
-        return [
-            "max_tokens",
-            "temperature",
-            "stream",
-        ]
+        supported_params = CohereChatConfig.get_supported_openai_params(
+            self, model=model
+        )
+        return supported_params
 
     def map_openai_params(
         self,
@@ -68,11 +66,10 @@ class AmazonCohereConfig(AmazonInvokeConfig, BaseConfig):
         model: str,
         drop_params: bool,
     ) -> dict:
-        for k, v in non_default_params.items():
-            if k == "stream":
-                optional_params["stream"] = v
-            if k == "temperature":
-                optional_params["temperature"] = v
-            if k == "max_tokens":
-                optional_params["max_tokens"] = v
-        return optional_params
+        return CohereChatConfig.map_openai_params(
+            self,
+            non_default_params=non_default_params,
+            optional_params=optional_params,
+            model=model,
+            drop_params=drop_params,
+        )
diff --git a/litellm/llms/bedrock/chat/invoke_transformations/amazon_nova_transformation.py b/litellm/llms/bedrock/chat/invoke_transformations/amazon_nova_transformation.py
index 9d41beceff..a81d55f0ad 100644
--- a/litellm/llms/bedrock/chat/invoke_transformations/amazon_nova_transformation.py
+++ b/litellm/llms/bedrock/chat/invoke_transformations/amazon_nova_transformation.py
@@ -6,14 +6,21 @@ Inherits from `AmazonConverseConfig`
 Nova + Invoke API Tutorial: https://docs.aws.amazon.com/nova/latest/userguide/using-invoke-api.html
 """
 
-from typing import List
+from typing import Any, List, Optional
+
+import httpx
 
 import litellm
+from litellm.litellm_core_utils.litellm_logging import Logging
 from litellm.types.llms.bedrock import BedrockInvokeNovaRequest
 from litellm.types.llms.openai import AllMessageValues
+from litellm.types.utils import ModelResponse
+
+from ..converse_transformation import AmazonConverseConfig
+from .base_invoke_transformation import AmazonInvokeConfig
 
 
-class AmazonInvokeNovaConfig(litellm.AmazonConverseConfig):
+class AmazonInvokeNovaConfig(AmazonInvokeConfig, AmazonConverseConfig):
     """
     Config for sending `nova` requests to `/bedrock/invoke/`
     """
@@ -21,6 +28,20 @@ class AmazonInvokeNovaConfig(litellm.AmazonConverseConfig):
     def __init__(self, **kwargs):
         super().__init__(**kwargs)
 
+    def get_supported_openai_params(self, model: str) -> list:
+        return AmazonConverseConfig.get_supported_openai_params(self, model)
+
+    def map_openai_params(
+        self,
+        non_default_params: dict,
+        optional_params: dict,
+        model: str,
+        drop_params: bool,
+    ) -> dict:
+        return AmazonConverseConfig.map_openai_params(
+            self, non_default_params, optional_params, model, drop_params
+        )
+
     def transform_request(
         self,
         model: str,
@@ -29,7 +50,8 @@ class AmazonInvokeNovaConfig(litellm.AmazonConverseConfig):
         litellm_params: dict,
         headers: dict,
     ) -> dict:
-        _transformed_nova_request = super().transform_request(
+        _transformed_nova_request = AmazonConverseConfig.transform_request(
+            self,
             model=model,
             messages=messages,
             optional_params=optional_params,
@@ -45,6 +67,35 @@ class AmazonInvokeNovaConfig(litellm.AmazonConverseConfig):
         )
         return bedrock_invoke_nova_request
 
+    def transform_response(
+        self,
+        model: str,
+        raw_response: httpx.Response,
+        model_response: ModelResponse,
+        logging_obj: Logging,
+        request_data: dict,
+        messages: List[AllMessageValues],
+        optional_params: dict,
+        litellm_params: dict,
+        encoding: Any,
+        api_key: Optional[str] = None,
+        json_mode: Optional[bool] = None,
+    ) -> litellm.ModelResponse:
+        return AmazonConverseConfig.transform_response(
+            self,
+            model,
+            raw_response,
+            model_response,
+            logging_obj,
+            request_data,
+            messages,
+            optional_params,
+            litellm_params,
+            encoding,
+            api_key,
+            json_mode,
+        )
+
     def _filter_allowed_fields(
         self, bedrock_invoke_nova_request: BedrockInvokeNovaRequest
     ) -> dict:
diff --git a/litellm/llms/custom_httpx/aiohttp_handler.py b/litellm/llms/custom_httpx/aiohttp_handler.py
index 12f4d962dc..72092cf261 100644
--- a/litellm/llms/custom_httpx/aiohttp_handler.py
+++ b/litellm/llms/custom_httpx/aiohttp_handler.py
@@ -218,6 +218,10 @@ class BaseLLMAIOHTTPHandler:
         provider_config = ProviderConfigManager.get_provider_chat_config(
             model=model, provider=litellm.LlmProviders(custom_llm_provider)
         )
+        if provider_config is None:
+            raise ValueError(
+                f"Provider config not found for model: {model} and provider: {custom_llm_provider}"
+            )
         # get config from model, custom llm provider
         headers = provider_config.validate_environment(
             api_key=api_key,
diff --git a/litellm/llms/custom_httpx/llm_http_handler.py b/litellm/llms/custom_httpx/llm_http_handler.py
index 66324e840e..5778f0228f 100644
--- a/litellm/llms/custom_httpx/llm_http_handler.py
+++ b/litellm/llms/custom_httpx/llm_http_handler.py
@@ -234,6 +234,10 @@ class BaseLLMHTTPHandler:
         provider_config = ProviderConfigManager.get_provider_chat_config(
             model=model, provider=litellm.LlmProviders(custom_llm_provider)
         )
+        if provider_config is None:
+            raise ValueError(
+                f"Provider config not found for model: {model} and provider: {custom_llm_provider}"
+            )
 
         # get config from model, custom llm provider
         headers = provider_config.validate_environment(
diff --git a/litellm/llms/litellm_proxy/chat/transformation.py b/litellm/llms/litellm_proxy/chat/transformation.py
index dadd921ab8..80774ccd55 100644
--- a/litellm/llms/litellm_proxy/chat/transformation.py
+++ b/litellm/llms/litellm_proxy/chat/transformation.py
@@ -10,6 +10,11 @@ from ...openai.chat.gpt_transformation import OpenAIGPTConfig
 
 
 class LiteLLMProxyChatConfig(OpenAIGPTConfig):
+    def get_supported_openai_params(self, model: str) -> List:
+        list = super().get_supported_openai_params(model)
+        list.append("thinking")
+        return list
+
     def _get_openai_compatible_provider_info(
         self, api_base: Optional[str], api_key: Optional[str]
     ) -> Tuple[Optional[str], Optional[str]]:
diff --git a/litellm/llms/vertex_ai/common_utils.py b/litellm/llms/vertex_ai/common_utils.py
index 0ff42806be..be8e4749dd 100644
--- a/litellm/llms/vertex_ai/common_utils.py
+++ b/litellm/llms/vertex_ai/common_utils.py
@@ -6,6 +6,7 @@ import httpx
 import litellm
 from litellm import supports_response_schema, supports_system_messages, verbose_logger
 from litellm.constants import DEFAULT_MAX_RECURSE_DEPTH
+from litellm.litellm_core_utils.prompt_templates.common_utils import unpack_defs
 from litellm.llms.base_llm.chat.transformation import BaseLLMException
 from litellm.types.llms.vertex_ai import PartType, Schema
 
@@ -229,39 +230,6 @@ def filter_schema_fields(
     return result
 
 
-def unpack_defs(schema, defs):
-    properties = schema.get("properties", None)
-    if properties is None:
-        return
-
-    for name, value in properties.items():
-        ref_key = value.get("$ref", None)
-        if ref_key is not None:
-            ref = defs[ref_key.split("defs/")[-1]]
-            unpack_defs(ref, defs)
-            properties[name] = ref
-            continue
-
-        anyof = value.get("anyOf", None)
-        if anyof is not None:
-            for i, atype in enumerate(anyof):
-                ref_key = atype.get("$ref", None)
-                if ref_key is not None:
-                    ref = defs[ref_key.split("defs/")[-1]]
-                    unpack_defs(ref, defs)
-                    anyof[i] = ref
-            continue
-
-        items = value.get("items", None)
-        if items is not None:
-            ref_key = items.get("$ref", None)
-            if ref_key is not None:
-                ref = defs[ref_key.split("defs/")[-1]]
-                unpack_defs(ref, defs)
-                value["items"] = ref
-                continue
-
-
 def convert_anyof_null_to_nullable(schema, depth=0):
     if depth > DEFAULT_MAX_RECURSE_DEPTH:
         raise ValueError(
diff --git a/litellm/llms/vertex_ai/vertex_ai_partner_models/ai21/transformation.py b/litellm/llms/vertex_ai/vertex_ai_partner_models/ai21/transformation.py
index d87b2e0311..8ffc00cc95 100644
--- a/litellm/llms/vertex_ai/vertex_ai_partner_models/ai21/transformation.py
+++ b/litellm/llms/vertex_ai/vertex_ai_partner_models/ai21/transformation.py
@@ -2,9 +2,10 @@ import types
 from typing import Optional
 
 import litellm
+from litellm.llms.openai.chat.gpt_transformation import OpenAIGPTConfig
 
 
-class VertexAIAi21Config:
+class VertexAIAi21Config(OpenAIGPTConfig):
     """
     Reference: https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/ai21
 
@@ -40,9 +41,6 @@ class VertexAIAi21Config:
             and v is not None
         }
 
-    def get_supported_openai_params(self):
-        return litellm.OpenAIConfig().get_supported_openai_params(model="gpt-3.5-turbo")
-
     def map_openai_params(
         self,
         non_default_params: dict,
diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index e345815fb2..a76356a3aa 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -2300,6 +2300,18 @@
         "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/000-000.mistral-ai-large-2407-offer?tab=Overview",
         "supports_tool_choice": true
     },
+    "azure_ai/mistral-large-latest": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000002,
+        "output_cost_per_token": 0.000006,
+        "litellm_provider": "azure_ai",
+        "supports_function_calling": true,
+        "mode": "chat",
+        "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/000-000.mistral-ai-large-2407-offer?tab=Overview",
+        "supports_tool_choice": true
+    },
     "azure_ai/ministral-3b": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,
@@ -8444,7 +8456,8 @@
         "input_cost_per_token": 0.0000015,
         "output_cost_per_token": 0.0000020,
         "litellm_provider": "bedrock",
-        "mode": "chat"
+        "mode": "chat",
+        "supports_tool_choice": true
     },
     "bedrock/*/1-month-commitment/cohere.command-text-v14": {
         "max_tokens": 4096, 
@@ -8471,7 +8484,8 @@
         "input_cost_per_token": 0.0000003,
         "output_cost_per_token": 0.0000006,
         "litellm_provider": "bedrock",
-        "mode": "chat"
+        "mode": "chat",
+        "supports_tool_choice": true
     },
     "bedrock/*/1-month-commitment/cohere.command-light-text-v14": {
         "max_tokens": 4096, 
@@ -8498,7 +8512,8 @@
         "input_cost_per_token": 0.0000030,
         "output_cost_per_token": 0.000015,
         "litellm_provider": "bedrock",
-        "mode": "chat"
+        "mode": "chat",
+        "supports_tool_choice": true
     },
     "cohere.command-r-v1:0": {
         "max_tokens": 4096, 
@@ -8507,7 +8522,8 @@
         "input_cost_per_token": 0.0000005,
         "output_cost_per_token": 0.0000015,
         "litellm_provider": "bedrock",
-        "mode": "chat"
+        "mode": "chat",
+        "supports_tool_choice": true
     },
     "cohere.embed-english-v3": {
         "max_tokens": 512, 
diff --git a/litellm/utils.py b/litellm/utils.py
index 7ea07e4b0b..f807990f60 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -3376,7 +3376,6 @@ def get_optional_params(  # noqa: PLR0915
                     if drop_params is not None and isinstance(drop_params, bool)
                     else False
                 ),
-                messages=messages,
             )
 
         elif "anthropic" in bedrock_base_model and bedrock_route == "invoke":
@@ -3719,6 +3718,17 @@ def get_optional_params(  # noqa: PLR0915
                     else False
                 ),
             )
+    elif provider_config is not None:
+        optional_params = provider_config.map_openai_params(
+            non_default_params=non_default_params,
+            optional_params=optional_params,
+            model=model,
+            drop_params=(
+                drop_params
+                if drop_params is not None and isinstance(drop_params, bool)
+                else False
+            ),
+        )
     else:  # assume passing in params for openai-like api
         optional_params = litellm.OpenAILikeChatConfig().map_openai_params(
             non_default_params=non_default_params,
@@ -6202,7 +6212,7 @@ class ProviderConfigManager:
     @staticmethod
     def get_provider_chat_config(  # noqa: PLR0915
         model: str, provider: LlmProviders
-    ) -> BaseConfig:
+    ) -> Optional[BaseConfig]:
         """
         Returns the provider config for a given provider.
         """
@@ -6233,9 +6243,22 @@ class ProviderConfigManager:
             return litellm.AnthropicConfig()
         elif litellm.LlmProviders.ANTHROPIC_TEXT == provider:
             return litellm.AnthropicTextConfig()
+        elif litellm.LlmProviders.VERTEX_AI_BETA == provider:
+            return litellm.VertexGeminiConfig()
         elif litellm.LlmProviders.VERTEX_AI == provider:
-            if "claude" in model:
+            if "gemini" in model:
+                return litellm.VertexGeminiConfig()
+            elif "claude" in model:
                 return litellm.VertexAIAnthropicConfig()
+            elif model in litellm.vertex_mistral_models:
+                if "codestral" in model:
+                    return litellm.CodestralTextCompletionConfig()
+                else:
+                    return litellm.MistralConfig()
+            elif model in litellm.vertex_ai_ai21_models:
+                return litellm.VertexAIAi21Config()
+            else:  # use generic openai-like param mapping
+                return litellm.VertexAILlama3Config()
         elif litellm.LlmProviders.CLOUDFLARE == provider:
             return litellm.CloudflareChatConfig()
         elif litellm.LlmProviders.SAGEMAKER_CHAT == provider:
@@ -6258,7 +6281,6 @@ class ProviderConfigManager:
             litellm.LlmProviders.CUSTOM == provider
             or litellm.LlmProviders.CUSTOM_OPENAI == provider
             or litellm.LlmProviders.OPENAI_LIKE == provider
-            or litellm.LlmProviders.LITELLM_PROXY == provider
         ):
             return litellm.OpenAILikeChatConfig()
         elif litellm.LlmProviders.AIOHTTP_OPENAI == provider:
@@ -6363,9 +6385,15 @@ class ProviderConfigManager:
                 return litellm.AmazonMistralConfig()
             elif bedrock_invoke_provider == "deepseek_r1":  # deepseek models on bedrock
                 return litellm.AmazonDeepSeekR1Config()
+            elif bedrock_invoke_provider == "nova":
+                return litellm.AmazonInvokeNovaConfig()
             else:
                 return litellm.AmazonInvokeConfig()
-        return litellm.OpenAIGPTConfig()
+        elif litellm.LlmProviders.LITELLM_PROXY == provider:
+            return litellm.LiteLLMProxyChatConfig()
+        elif litellm.LlmProviders.OPENAI == provider:
+            return litellm.OpenAIGPTConfig()
+        return None
 
     @staticmethod
     def get_provider_embedding_config(
diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json
index e345815fb2..a76356a3aa 100644
--- a/model_prices_and_context_window.json
+++ b/model_prices_and_context_window.json
@@ -2300,6 +2300,18 @@
         "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/000-000.mistral-ai-large-2407-offer?tab=Overview",
         "supports_tool_choice": true
     },
+    "azure_ai/mistral-large-latest": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000002,
+        "output_cost_per_token": 0.000006,
+        "litellm_provider": "azure_ai",
+        "supports_function_calling": true,
+        "mode": "chat",
+        "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/000-000.mistral-ai-large-2407-offer?tab=Overview",
+        "supports_tool_choice": true
+    },
     "azure_ai/ministral-3b": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,
@@ -8444,7 +8456,8 @@
         "input_cost_per_token": 0.0000015,
         "output_cost_per_token": 0.0000020,
         "litellm_provider": "bedrock",
-        "mode": "chat"
+        "mode": "chat",
+        "supports_tool_choice": true
     },
     "bedrock/*/1-month-commitment/cohere.command-text-v14": {
         "max_tokens": 4096, 
@@ -8471,7 +8484,8 @@
         "input_cost_per_token": 0.0000003,
         "output_cost_per_token": 0.0000006,
         "litellm_provider": "bedrock",
-        "mode": "chat"
+        "mode": "chat",
+        "supports_tool_choice": true
     },
     "bedrock/*/1-month-commitment/cohere.command-light-text-v14": {
         "max_tokens": 4096, 
@@ -8498,7 +8512,8 @@
         "input_cost_per_token": 0.0000030,
         "output_cost_per_token": 0.000015,
         "litellm_provider": "bedrock",
-        "mode": "chat"
+        "mode": "chat",
+        "supports_tool_choice": true
     },
     "cohere.command-r-v1:0": {
         "max_tokens": 4096, 
@@ -8507,7 +8522,8 @@
         "input_cost_per_token": 0.0000005,
         "output_cost_per_token": 0.0000015,
         "litellm_provider": "bedrock",
-        "mode": "chat"
+        "mode": "chat",
+        "supports_tool_choice": true
     },
     "cohere.embed-english-v3": {
         "max_tokens": 512, 
diff --git a/tests/litellm_utils_tests/log.txt b/tests/litellm_utils_tests/log.txt
new file mode 100644
index 0000000000..bce7b79031
--- /dev/null
+++ b/tests/litellm_utils_tests/log.txt
@@ -0,0 +1,24 @@
+============================= test session starts ==============================
+platform darwin -- Python 3.13.1, pytest-8.3.5, pluggy-1.5.0 -- /Users/krrishdholakia/Documents/litellm/myenv/bin/python3.13
+cachedir: .pytest_cache
+rootdir: /Users/krrishdholakia/Documents/litellm
+configfile: pyproject.toml
+plugins: respx-0.22.0, postgresql-7.0.1, anyio-4.4.0, asyncio-0.26.0, mock-3.14.0, ddtrace-2.19.0rc1
+asyncio: mode=Mode.STRICT, asyncio_default_fixture_loop_scope=None, asyncio_default_test_loop_scope=function
+collecting ... collected 3 items
+
+test_supports_tool_choice.py::test_check_provider_match PASSED           [ 33%]
+test_supports_tool_choice.py::test_supports_tool_choice PASSED           [ 66%]
+test_supports_tool_choice.py::test_supports_tool_choice_simple_tests PASSED [100%]
+
+=============================== warnings summary ===============================
+../../myenv/lib/python3.13/site-packages/pydantic/_internal/_config.py:295
+  /Users/krrishdholakia/Documents/litellm/myenv/lib/python3.13/site-packages/pydantic/_internal/_config.py:295: PydanticDeprecatedSince20: Support for class-based `config` is deprecated, use ConfigDict instead. Deprecated in Pydantic V2.0 to be removed in V3.0. See Pydantic V2 Migration Guide at https://errors.pydantic.dev/2.10/migration/
+    warnings.warn(DEPRECATION_MESSAGE, DeprecationWarning)
+
+../../litellm/caching/llm_caching_handler.py:17
+  /Users/krrishdholakia/Documents/litellm/litellm/caching/llm_caching_handler.py:17: DeprecationWarning: There is no current event loop
+    event_loop = asyncio.get_event_loop()
+
+-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
+======================== 3 passed, 2 warnings in 0.92s =========================
diff --git a/tests/litellm_utils_tests/test_supports_tool_choice.py b/tests/litellm_utils_tests/test_supports_tool_choice.py
index e09be15319..d9e4f0ca82 100644
--- a/tests/litellm_utils_tests/test_supports_tool_choice.py
+++ b/tests/litellm_utils_tests/test_supports_tool_choice.py
@@ -117,9 +117,9 @@ async def test_supports_tool_choice():
     """
     # Load model prices
     litellm._turn_on_debug()
-    local_path = "../../model_prices_and_context_window.json"
-    prod_path = "./model_prices_and_context_window.json"
-    with open(prod_path, "r") as f:
+    # path = "../../model_prices_and_context_window.json"
+    path = "./model_prices_and_context_window.json"
+    with open(path, "r") as f:
         model_prices = json.load(f)
     litellm.model_cost = model_prices
     config_manager = ProviderConfigManager()
@@ -137,8 +137,10 @@ async def test_supports_tool_choice():
             or model_name in block_list
             or "azure/eu" in model_name
             or "azure/us" in model_name
+            or "codestral" in model_name
             or "o1" in model_name
             or "o3" in model_name
+            or "mistral" in model_name
         ):
             continue
 
@@ -152,8 +154,11 @@ async def test_supports_tool_choice():
         print("LLM provider", provider)
         provider_enum = LlmProviders(provider)
         config = config_manager.get_provider_chat_config(model, provider_enum)
-        supported_params = config.get_supported_openai_params(model)
-        print("supported_params", supported_params)
+        if config:
+            supported_params = config.get_supported_openai_params(model)
+            print("supported_params", supported_params)
+        else:
+            raise Exception(f"No config found for {model_name}, provider: {provider}")
 
         # Check tool_choice support
         supports_tool_choice_result = litellm.utils.supports_tool_choice(
@@ -162,7 +167,5 @@ async def test_supports_tool_choice():
         tool_choice_in_params = "tool_choice" in supported_params
 
         assert supports_tool_choice_result == tool_choice_in_params, (
-            f"Tool choice support mismatch for {model_name}:\n"
-            f"supports_tool_choice() returned: {supports_tool_choice_result}\n"
-            f"tool_choice in supported params: {tool_choice_in_params}"
+            f"Tool choice support mismatch for {model_name}. supports_tool_choice() returned: {supports_tool_choice_result}, tool_choice in supported params: {tool_choice_in_params}"
         )
diff --git a/tests/llm_translation/base_llm_unit_tests.py b/tests/llm_translation/base_llm_unit_tests.py
index a8400cf72e..9e21260410 100644
--- a/tests/llm_translation/base_llm_unit_tests.py
+++ b/tests/llm_translation/base_llm_unit_tests.py
@@ -474,6 +474,7 @@ class BaseLLMChatTest(ABC):
         """
         PROD Test: ensure nested json schema sent to proxy works as expected.
         """
+        litellm._turn_on_debug()
         from pydantic import BaseModel
         from litellm.utils import supports_response_schema
         from litellm.llms.base_llm.base_utils import type_to_response_format_param
diff --git a/tests/llm_translation/test_max_completion_tokens.py b/tests/llm_translation/test_max_completion_tokens.py
index 198cd45a55..9bab536467 100644
--- a/tests/llm_translation/test_max_completion_tokens.py
+++ b/tests/llm_translation/test_max_completion_tokens.py
@@ -150,9 +150,11 @@ def test_all_model_configs():
         {"max_completion_tokens": 10}, {}, "llama3", drop_params=False
     ) == {"max_tokens": 10}
 
-    assert "max_completion_tokens" in VertexAIAi21Config().get_supported_openai_params()
+    assert "max_completion_tokens" in VertexAIAi21Config().get_supported_openai_params(
+        model="jamba-1.5-mini@001"
+    )
     assert VertexAIAi21Config().map_openai_params(
-        {"max_completion_tokens": 10}, {}, "llama3", drop_params=False
+        {"max_completion_tokens": 10}, {}, "jamba-1.5-mini@001", drop_params=False
     ) == {"max_tokens": 10}
 
     from litellm.llms.fireworks_ai.chat.transformation import (
diff --git a/tests/llm_translation/test_optional_params.py b/tests/llm_translation/test_optional_params.py
index 8180cc7279..078a0beef0 100644
--- a/tests/llm_translation/test_optional_params.py
+++ b/tests/llm_translation/test_optional_params.py
@@ -1404,6 +1404,13 @@ def test_azure_modalities_param():
     assert optional_params["modalities"] == ["text", "audio"]
     assert optional_params["audio"] == {"type": "audio_input", "input": "test.wav"}
 
+def test_litellm_proxy_thinking_param():
+    optional_params = get_optional_params(
+        model="gpt-4o",
+        custom_llm_provider="litellm_proxy",
+        thinking={"type": "enabled", "budget_tokens": 1024},
+    )
+    assert optional_params["thinking"] == {"type": "enabled", "budget_tokens": 1024}
 
 def test_gemini_modalities_param():
     optional_params = get_optional_params(
@@ -1441,3 +1448,4 @@ def test_anthropic_unified_reasoning_content(model, provider):
         reasoning_effort="high",
     )
     assert optional_params["thinking"] == {"type": "enabled", "budget_tokens": 4096}
+
diff --git a/tests/local_testing/test_completion.py b/tests/local_testing/test_completion.py
index 3ec7b01b49..ed07331f1d 100644
--- a/tests/local_testing/test_completion.py
+++ b/tests/local_testing/test_completion.py
@@ -880,6 +880,9 @@ def test_completion_azure_mistral_large_function_calling(provider):
     This primarily tests if the 'Function()' pydantic object correctly handles argument param passed in as a dict vs. string
     """
     litellm.set_verbose = True
+    os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
+
+    model_cost = litellm.get_model_cost_map(url="")
     tools = [
         {
             "type": "function",
@@ -1903,16 +1906,16 @@ def test_completion_openai():
 @pytest.mark.parametrize(
     "model, api_version",
     [
-        ("gpt-4o-2024-08-06", None),
-        ("azure/chatgpt-v-2", None),
+        # ("gpt-4o-2024-08-06", None),
+        # ("azure/chatgpt-v-2", None),
         ("bedrock/anthropic.claude-3-sonnet-20240229-v1:0", None),
-        ("azure/gpt-4o", "2024-08-01-preview"),
+        # ("azure/gpt-4o", "2024-08-01-preview"),
     ],
 )
 @pytest.mark.flaky(retries=3, delay=1)
 def test_completion_openai_pydantic(model, api_version):
     try:
-        litellm.set_verbose = True
+        litellm._turn_on_debug()
         from pydantic import BaseModel
 
         messages = [