diff --git a/skyvern/forge/forge_app.py b/skyvern/forge/forge_app.py
index 4b91c72a..69ad945e 100644
--- a/skyvern/forge/forge_app.py
+++ b/skyvern/forge/forge_app.py
@@ -12,8 +12,8 @@ from skyvern.forge.agent_functions import AgentFunction
 from skyvern.forge.forge_openai_client import ForgeAsyncHttpxClientWrapper
 from skyvern.forge.sdk.api.azure import AzureClientFactory
 from skyvern.forge.sdk.api.custom_credential_client import CustomCredentialAPIClient
+from skyvern.forge.sdk.api.llm.api_handler import LLMAPIHandler
 from skyvern.forge.sdk.api.llm.api_handler_factory import LLMAPIHandlerFactory
-from skyvern.forge.sdk.api.llm.models import LLMAPIHandler
 from skyvern.forge.sdk.api.real_azure import RealAzureClientFactory
 from skyvern.forge.sdk.artifact.manager import ArtifactManager
 from skyvern.forge.sdk.artifact.storage.base import BaseStorage
diff --git a/skyvern/forge/sdk/api/llm/api_handler.py b/skyvern/forge/sdk/api/llm/api_handler.py
new file mode 100644
index 00000000..1d68afb5
--- /dev/null
+++ b/skyvern/forge/sdk/api/llm/api_handler.py
@@ -0,0 +1,45 @@
+from typing import Any, Awaitable, Protocol
+
+from skyvern.forge.sdk.models import Step
+from skyvern.forge.sdk.schemas.ai_suggestions import AISuggestion
+from skyvern.forge.sdk.schemas.task_v2 import TaskV2, Thought
+from skyvern.utils.image_resizer import Resolution
+
+
+class LLMAPIHandler(Protocol):
+    def __call__(
+        self,
+        prompt: str,
+        prompt_name: str,
+        step: Step | None = None,
+        task_v2: TaskV2 | None = None,
+        thought: Thought | None = None,
+        ai_suggestion: AISuggestion | None = None,
+        screenshots: list[bytes] | None = None,
+        parameters: dict[str, Any] | None = None,
+        organization_id: str | None = None,
+        tools: list | None = None,
+        use_message_history: bool = False,
+        raw_response: bool = False,
+        window_dimension: Resolution | None = None,
+        force_dict: bool = True,
+    ) -> Awaitable[dict[str, Any] | Any]: ...
+
+
+async def dummy_llm_api_handler(
+    prompt: str,
+    prompt_name: str,
+    step: Step | None = None,
+    task_v2: TaskV2 | None = None,
+    thought: Thought | None = None,
+    ai_suggestion: AISuggestion | None = None,
+    screenshots: list[bytes] | None = None,
+    parameters: dict[str, Any] | None = None,
+    organization_id: str | None = None,
+    tools: list | None = None,
+    use_message_history: bool = False,
+    raw_response: bool = False,
+    window_dimension: Resolution | None = None,
+    force_dict: bool = True,
+) -> dict[str, Any] | Any:
+    raise NotImplementedError("Your LLM provider is not configured. Please configure it in the .env file.")
diff --git a/skyvern/forge/sdk/api/llm/api_handler_factory.py b/skyvern/forge/sdk/api/llm/api_handler_factory.py
index 635d63c5..db3ece1e 100644
--- a/skyvern/forge/sdk/api/llm/api_handler_factory.py
+++ b/skyvern/forge/sdk/api/llm/api_handler_factory.py
@@ -20,6 +20,7 @@ from skyvern.config import settings
 from skyvern.exceptions import SkyvernContextWindowExceededError
 from skyvern.forge import app
 from skyvern.forge.forge_openai_client import ForgeAsyncHttpxClientWrapper
+from skyvern.forge.sdk.api.llm.api_handler import LLMAPIHandler, dummy_llm_api_handler
 from skyvern.forge.sdk.api.llm.config_registry import LLMConfigRegistry
 from skyvern.forge.sdk.api.llm.exceptions import (
     DuplicateCustomLLMProviderError,
@@ -29,10 +30,8 @@ from skyvern.forge.sdk.api.llm.exceptions import (
 )
 from skyvern.forge.sdk.api.llm.models import (
     LLMAllowedFailsPolicy,
-    LLMAPIHandler,
     LLMConfig,
     LLMRouterConfig,
-    dummy_llm_api_handler,
 )
 from skyvern.forge.sdk.api.llm.ui_tars_response import UITarsResponse
 from skyvern.forge.sdk.api.llm.utils import llm_messages_builder, llm_messages_builder_with_history, parse_api_response
diff --git a/skyvern/forge/sdk/api/llm/models.py b/skyvern/forge/sdk/api/llm/models.py
index b60273be..faa399d8 100644
--- a/skyvern/forge/sdk/api/llm/models.py
+++ b/skyvern/forge/sdk/api/llm/models.py
@@ -1,11 +1,7 @@
 from dataclasses import dataclass, field
-from typing import Any, Awaitable, Literal, Optional, Protocol, TypedDict
+from typing import Any, Literal, Optional, TypedDict
 
-from skyvern.forge.sdk.models import Step
-from skyvern.forge.sdk.schemas.ai_suggestions import AISuggestion
-from skyvern.forge.sdk.schemas.task_v2 import TaskV2, Thought
 from skyvern.forge.sdk.settings_manager import SettingsManager
-from skyvern.utils.image_resizer import Resolution
 
 
 class LiteLLMParams(TypedDict, total=False):
@@ -93,42 +89,3 @@ class LLMRouterConfig(LLMConfigBase):
     max_completion_tokens: int | None = None
     reasoning_effort: str | None = None
     temperature: float | None = SettingsManager.get_settings().LLM_CONFIG_TEMPERATURE
-
-
-class LLMAPIHandler(Protocol):
-    def __call__(
-        self,
-        prompt: str,
-        prompt_name: str,
-        step: Step | None = None,
-        task_v2: TaskV2 | None = None,
-        thought: Thought | None = None,
-        ai_suggestion: AISuggestion | None = None,
-        screenshots: list[bytes] | None = None,
-        parameters: dict[str, Any] | None = None,
-        organization_id: str | None = None,
-        tools: list | None = None,
-        use_message_history: bool = False,
-        raw_response: bool = False,
-        window_dimension: Resolution | None = None,
-        force_dict: bool = True,
-    ) -> Awaitable[dict[str, Any] | Any]: ...
-
-
-async def dummy_llm_api_handler(
-    prompt: str,
-    prompt_name: str,
-    step: Step | None = None,
-    task_v2: TaskV2 | None = None,
-    thought: Thought | None = None,
-    ai_suggestion: AISuggestion | None = None,
-    screenshots: list[bytes] | None = None,
-    parameters: dict[str, Any] | None = None,
-    organization_id: str | None = None,
-    tools: list | None = None,
-    use_message_history: bool = False,
-    raw_response: bool = False,
-    window_dimension: Resolution | None = None,
-    force_dict: bool = True,
-) -> dict[str, Any] | Any:
-    raise NotImplementedError("Your LLM provider is not configured. Please configure it in the .env file.")
diff --git a/skyvern/forge/sdk/experimentation/llm_prompt_config.py b/skyvern/forge/sdk/experimentation/llm_prompt_config.py
index 5ba7c6de..39b89837 100644
--- a/skyvern/forge/sdk/experimentation/llm_prompt_config.py
+++ b/skyvern/forge/sdk/experimentation/llm_prompt_config.py
@@ -5,8 +5,8 @@ import json
 import structlog
 
 from skyvern.forge import app
+from skyvern.forge.sdk.api.llm.api_handler import LLMAPIHandler
 from skyvern.forge.sdk.api.llm.api_handler_factory import LLMAPIHandlerFactory
-from skyvern.forge.sdk.api.llm.models import LLMAPIHandler
 
 LOG = structlog.get_logger()
 
diff --git a/skyvern/forge/sdk/workflow/models/block.py b/skyvern/forge/sdk/workflow/models/block.py
index da1786f7..2df60151 100644
--- a/skyvern/forge/sdk/workflow/models/block.py
+++ b/skyvern/forge/sdk/workflow/models/block.py
@@ -56,8 +56,8 @@ from skyvern.forge.sdk.api.files import (
     download_from_s3,
     get_path_for_workflow_download_directory,
 )
+from skyvern.forge.sdk.api.llm.api_handler import LLMAPIHandler
 from skyvern.forge.sdk.api.llm.api_handler_factory import LLMAPIHandlerFactory
-from skyvern.forge.sdk.api.llm.models import LLMAPIHandler
 from skyvern.forge.sdk.artifact.models import ArtifactType
 from skyvern.forge.sdk.core import skyvern_context
 from skyvern.forge.sdk.core.aiohttp_helper import aiohttp_request