diff --git a/invokeai/app/api/routers/model_manager.py b/invokeai/app/api/routers/model_manager.py
index 78ef965d5a..58950d6692 100644
--- a/invokeai/app/api/routers/model_manager.py
+++ b/invokeai/app/api/routers/model_manager.py
@@ -75,7 +75,6 @@ example_model_config = {
     "description": "string",
     "source": "string",
     "last_modified": 0,
-    "vae": "string",
     "variant": "normal",
     "prediction_type": "epsilon",
     "repo_variant": "fp16",
diff --git a/invokeai/backend/model_manager/config.py b/invokeai/backend/model_manager/config.py
index 87c2feab5b..5aa07f44ee 100644
--- a/invokeai/backend/model_manager/config.py
+++ b/invokeai/backend/model_manager/config.py
@@ -260,7 +260,6 @@ class TextualInversionFolderConfig(ModelConfigBase):
 class _MainConfig(ModelConfigBase):
     """Model config for main models."""
 
-    vae: Optional[str] = Field(default=None)
     variant: ModelVariantType = ModelVariantType.Normal
     prediction_type: SchedulerPredictionType = SchedulerPredictionType.Epsilon
     upcast_attention: bool = False
diff --git a/invokeai/backend/model_manager/load/model_loader_registry.py b/invokeai/backend/model_manager/load/model_loader_registry.py
index e7747650d4..4b6a6ff57f 100644
--- a/invokeai/backend/model_manager/load/model_loader_registry.py
+++ b/invokeai/backend/model_manager/load/model_loader_registry.py
@@ -15,9 +15,7 @@ Use like this:
 
 """
 
-import hashlib
 from abc import ABC, abstractmethod
-from pathlib import Path
 from typing import Callable, Dict, Optional, Tuple, Type
 
 from ..config import (
@@ -27,8 +25,6 @@ from ..config import (
     ModelFormat,
     ModelType,
     SubModelType,
-    VaeCheckpointConfig,
-    VaeDiffusersConfig,
 )
 from . import ModelLoaderBase
 
@@ -90,33 +86,15 @@ class ModelLoaderRegistry:
         cls, config: AnyModelConfig, submodel_type: Optional[SubModelType]
     ) -> Tuple[Type[ModelLoaderBase], ModelConfigBase, Optional[SubModelType]]:
         """Get subclass of ModelLoaderBase registered to handle base and type."""
-        # We have to handle VAE overrides here because this will change the model type and the corresponding implementation returned
-        conf2, submodel_type = cls._handle_subtype_overrides(config, submodel_type)
 
-        key1 = cls._to_registry_key(conf2.base, conf2.type, conf2.format)  # for a specific base type
-        key2 = cls._to_registry_key(BaseModelType.Any, conf2.type, conf2.format)  # with wildcard Any
+        key1 = cls._to_registry_key(config.base, config.type, config.format)  # for a specific base type
+        key2 = cls._to_registry_key(BaseModelType.Any, config.type, config.format)  # with wildcard Any
         implementation = cls._registry.get(key1) or cls._registry.get(key2)
         if not implementation:
             raise NotImplementedError(
                 f"No subclass of LoadedModel is registered for base={config.base}, type={config.type}, format={config.format}"
             )
-        return implementation, conf2, submodel_type
-
-    @classmethod
-    def _handle_subtype_overrides(
-        cls, config: AnyModelConfig, submodel_type: Optional[SubModelType]
-    ) -> Tuple[ModelConfigBase, Optional[SubModelType]]:
-        if submodel_type == SubModelType.Vae and hasattr(config, "vae") and config.vae is not None:
-            model_path = Path(config.vae)
-            config_class = (
-                VaeCheckpointConfig if model_path.suffix in [".pt", ".safetensors", ".ckpt"] else VaeDiffusersConfig
-            )
-            hash = hashlib.md5(model_path.as_posix().encode("utf-8")).hexdigest()
-            new_conf = config_class(path=model_path.as_posix(), name=model_path.stem, base=config.base, key=hash)
-            submodel_type = None
-        else:
-            new_conf = config
-        return new_conf, submodel_type
+        return implementation, config, submodel_type
 
     @staticmethod
     def _to_registry_key(base: BaseModelType, type: ModelType, format: ModelFormat) -> str: