diff --git a/litellm/proxy/guardrails/guardrail_hooks/neuraltrust/__init__.py b/litellm/proxy/guardrails/guardrail_hooks/neuraltrust/__init__.py
index 7710965c932..1a91bfebcbb 100644
--- a/litellm/proxy/guardrails/guardrail_hooks/neuraltrust/__init__.py
+++ b/litellm/proxy/guardrails/guardrail_hooks/neuraltrust/__init__.py
@@ -20,7 +20,7 @@ def initialize_guardrail(litellm_params: LitellmParams, guardrail: Guardrail) ->
         unreachable_fallback=litellm_params.unreachable_fallback,
         timeout=litellm_params.timeout,
         streaming_transform_mode=litellm_params.streaming_transform_mode,
-        guardrail_name=guardrail.get("guardrail_name", ""),
+        guardrail_name=guardrail["guardrail_name"],
         event_hook=litellm_params.mode,
         default_on=litellm_params.default_on,
     )
diff --git a/litellm/proxy/guardrails/guardrail_hooks/neuraltrust/neuraltrust.py b/litellm/proxy/guardrails/guardrail_hooks/neuraltrust/neuraltrust.py
index ca8d98c9cfe..9b258201dfd 100644
--- a/litellm/proxy/guardrails/guardrail_hooks/neuraltrust/neuraltrust.py
+++ b/litellm/proxy/guardrails/guardrail_hooks/neuraltrust/neuraltrust.py
@@ -6,10 +6,10 @@ Calls TrustGuard POST /v1/evaluate on pre_call (input) and post_call (output).
 from __future__ import annotations
 
 import os
-from collections.abc import Mapping, Sequence
+from collections.abc import Callable, Mapping, Sequence
 from itertools import chain, product
 from types import MappingProxyType
-from typing import TYPE_CHECKING, Final, Literal
+from typing import TYPE_CHECKING, Final, Literal, Protocol
 
 import httpx
 from fastapi import HTTPException
@@ -20,20 +20,20 @@ from litellm.exceptions import Timeout
 from litellm.integrations.custom_guardrail import (
     CustomGuardrail,
     get_session_id_from_request_data,
-    log_guardrail_information,
-)
-from litellm.llms.custom_httpx.http_handler import (
-    get_async_httpx_client,
-    httpxSpecialProvider,
 )
+from litellm.llms.custom_httpx import http_handler
+from litellm.llms.custom_httpx.http_handler import httpxSpecialProvider
 from litellm.types.guardrails import GuardrailEventHooks, Mode
 from litellm.types.llms.openai import AllMessageValues
-from litellm.types.proxy.guardrails.guardrail_hooks.neuraltrust import DEFAULT_API_BASE, DEFAULT_TIMEOUT
+from litellm.types.proxy.guardrails.guardrail_hooks.neuraltrust import (
+    DEFAULT_API_BASE,
+    DEFAULT_TIMEOUT,
+    NeuralTrustGuardrailConfigModel,
+)
 from litellm.types.utils import ChatCompletionMessageToolCall, GenericGuardrailAPIInputs
 
 if TYPE_CHECKING:
     from litellm.litellm_core_utils.litellm_logging import Logging as LiteLLMLoggingObj
-    from litellm.types.proxy.guardrails.guardrail_hooks.base import GuardrailConfigModel
 
 EVALUATE_PATH: Final = "/v1/evaluate"
 CONSUMER_ID_KEYS: Final = (
@@ -55,6 +55,17 @@ UNREACHABLE_HTTP_STATUSES: Final = frozenset({502, 504})
 TRANSFORM_MISSING: Final = "TrustGuard transform missing payload"
 
 
+class _TrustGuardClient(Protocol):
+    async def post(
+        self,
+        url: str,
+        *,
+        json: dict[str, object],
+        headers: dict[str, str],
+        timeout: float | httpx.Timeout | None,
+    ) -> httpx.Response: ...
+
+
 class _TrustGuardUnreachable(Exception):
     """Transport or availability failure; eligible for unreachable_fallback."""
 
@@ -192,11 +203,7 @@ class NeuralTrustGuardrail(CustomGuardrail):
     """LiteLLM hook that evaluates prompts and completions with TrustGuard."""
 
     @staticmethod
-    def get_config_model() -> type[GuardrailConfigModel]:
-        from litellm.types.proxy.guardrails.guardrail_hooks.neuraltrust import (
-            NeuralTrustGuardrailConfigModel,
-        )
-
+    def get_config_model() -> type[NeuralTrustGuardrailConfigModel]:
         return NeuralTrustGuardrailConfigModel
 
     @classmethod
@@ -218,9 +225,11 @@ class NeuralTrustGuardrail(CustomGuardrail):
         event_hook: GuardrailEventHooks | Mode | str | Sequence[str] | None = None,
         default_on: bool | None = None,
     ) -> None:
-        self.async_handler = get_async_httpx_client(
-            llm_provider=httpxSpecialProvider.GuardrailCallback,
+        factory_adapter: Final[TypeAdapter[Callable[[httpxSpecialProvider], _TrustGuardClient]]] = TypeAdapter(
+            Callable[[httpxSpecialProvider], _TrustGuardClient]
         )
+        client_factory: Final = factory_adapter.validate_python(getattr(http_handler, "get_async_httpx_client"))
+        self.async_handler = client_factory(httpxSpecialProvider.GuardrailCallback)
         self.api_base = (api_base or os.environ.get("TRUSTGUARD_API_BASE") or DEFAULT_API_BASE).rstrip("/")
         self.api_key = api_key or os.environ.get("TRUSTGUARD_API_KEY") or ""
         if not self.api_key:
@@ -245,7 +254,6 @@ class NeuralTrustGuardrail(CustomGuardrail):
             default_on=bool(default_on),
         )
 
-    @log_guardrail_information
     async def apply_guardrail(
         self,
         inputs: GenericGuardrailAPIInputs,
diff --git a/litellm/types/proxy/guardrails/guardrail_hooks/neuraltrust.py b/litellm/types/proxy/guardrails/guardrail_hooks/neuraltrust.py
index 73f7db0299a..42e6d221d8e 100644
--- a/litellm/types/proxy/guardrails/guardrail_hooks/neuraltrust.py
+++ b/litellm/types/proxy/guardrails/guardrail_hooks/neuraltrust.py
@@ -1,6 +1,6 @@
 from typing import Final, Literal
 
-from pydantic import Field
+from pydantic import BaseModel, Field
 
 from .base import GuardrailConfigModel
 
@@ -8,7 +8,7 @@ DEFAULT_API_BASE: Final = "https://trustguard.neuraltrust.ai"
 DEFAULT_TIMEOUT: Final = 5.0
 
 
-class NeuralTrustGuardrailConfigModel(GuardrailConfigModel):
+class NeuralTrustGuardrailConfigModel(GuardrailConfigModel[BaseModel]):
     """Config for the NeuralTrust TrustGuard native LiteLLM hook."""
 
     api_key: str | None = Field(