LoRA patching optimization (#6439)

* allow model patcher to optimize away the unpatching step when feasible

* remove lazy_offloading functionality

* allow model patcher to optimize away the unpatching step when feasible

* remove lazy_offloading functionality

* do not save original weights if there is a CPU copy of state dict

* Update invokeai/backend/model_manager/load/load_base.py

Co-authored-by: Ryan Dick <ryanjdick3@gmail.com>

* documentation fixes added during penultimate review

---------

Co-authored-by: Lincoln Stein <lstein@gmail.com>
Co-authored-by: Kent Keirsey <31807370+hipsterusername@users.noreply.github.com>
Co-authored-by: Ryan Dick <ryanjdick3@gmail.com>

docs/contributing/MODEL_MANAGER.md

@@ -1367,12 +1367,20 @@ the in-memory loaded model:
 | `model`        | AnyModel               | The instantiated model (details below) |
 | `locker`       | ModelLockerBase        | A context manager that mediates the movement of the model into VRAM |
 
-Because the loader can return multiple model types, it is typed to
+### get_model_by_key(key, [submodel]) -> LoadedModel
-return `AnyModel`, a Union `ModelMixin`, `torch.nn.Module`,
+
-`IAIOnnxRuntimeModel`, `IPAdapter`, `IPAdapterPlus`, and
+The `get_model_by_key()` method will retrieve the model using its
-`EmbeddingModelRaw`. `ModelMixin` is the base class of all diffusers
+unique database key. For example:
-models, `EmbeddingModelRaw` is used for LoRA and TextualInversion
+
-models. The others are obvious.
+loaded_model = loader.get_model_by_key('f13dd932c0c35c22dcb8d6cda4203764', SubModelType('vae'))
+
+`get_model_by_key()` may raise any of the following exceptions:
+
+* `UnknownModelException`   -- key not in database
+* `ModelNotFoundException`  -- key in database but model not found at path
+* `NotImplementedException` -- the loader doesn't know how to load this type of model
+
+### Using the Loaded Model in Inference
 
 `LoadedModel` acts as a context manager. The context loads the model
 into the execution device (e.g. VRAM on CUDA systems), locks the model
@@ -1380,16 +1388,32 @@ in the execution device for the duration of the context, and returns
 the model. Use it like this:
 
 ```
-model_info = loader.get_model_by_key('f13dd932c0c35c22dcb8d6cda4203764', SubModelType('vae'))
+loaded_model_= loader.get_model_by_key('f13dd932c0c35c22dcb8d6cda4203764', SubModelType('vae'))
-with model_info as vae:
+with loaded_model as vae:
  image = vae.decode(latents)[0]
 ```
 
-`get_model_by_key()` may raise any of the following exceptions:
+The object returned by the LoadedModel context manager is an
+`AnyModel`, which is a Union of `ModelMixin`, `torch.nn.Module`,
+`IAIOnnxRuntimeModel`, `IPAdapter`, `IPAdapterPlus`, and
+`EmbeddingModelRaw`. `ModelMixin` is the base class of all diffusers
+models, `EmbeddingModelRaw` is used for LoRA and TextualInversion
+models. The others are obvious.
+
+In addition, you may call `LoadedModel.model_on_device()`, a context
+manager that returns a tuple of the model's state dict in CPU and the
+model itself in VRAM. It is used to optimize the LoRA patching and
+unpatching process:
+
+```
+loaded_model_= loader.get_model_by_key('f13dd932c0c35c22dcb8d6cda4203764', SubModelType('vae'))
+with loaded_model.model_on_device() as (state_dict, vae):
+ image = vae.decode(latents)[0]
+```
+
+Since not all models have state dicts, the `state_dict` return value
+can be None.
 
-* `UnknownModelException`   -- key not in database
-* `ModelNotFoundException`  -- key in database but model not found at path
-* `NotImplementedException` -- the loader doesn't know how to load this type of model
 
 ### Emitting model loading events
 

invokeai/app/invocations/compel.py

@@ -81,9 +81,13 @@ class CompelInvocation(BaseInvocation):
 
         with (
             # apply all patches while the model is on the target device
-            text_encoder_info as text_encoder,
+            text_encoder_info.model_on_device() as (model_state_dict, text_encoder),
             tokenizer_info as tokenizer,
-            ModelPatcher.apply_lora_text_encoder(text_encoder, _lora_loader()),
+            ModelPatcher.apply_lora_text_encoder(
+                text_encoder,
+                loras=_lora_loader(),
+                model_state_dict=model_state_dict,
+            ),
             # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
             ModelPatcher.apply_clip_skip(text_encoder, self.clip.skipped_layers),
             ModelPatcher.apply_ti(tokenizer, text_encoder, ti_list) as (
@@ -172,9 +176,14 @@ class SDXLPromptInvocationBase:
 
         with (
             # apply all patches while the model is on the target device
-            text_encoder_info as text_encoder,
+            text_encoder_info.model_on_device() as (state_dict, text_encoder),
             tokenizer_info as tokenizer,
-            ModelPatcher.apply_lora(text_encoder, _lora_loader(), lora_prefix),
+            ModelPatcher.apply_lora(
+                text_encoder,
+                loras=_lora_loader(),
+                prefix=lora_prefix,
+                model_state_dict=state_dict,
+            ),
             # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
             ModelPatcher.apply_clip_skip(text_encoder, clip_field.skipped_layers),
             ModelPatcher.apply_ti(tokenizer, text_encoder, ti_list) as (

invokeai/app/invocations/latent.py

@@ -952,11 +952,15 @@ class DenoiseLatentsInvocation(BaseInvocation):
             assert isinstance(unet_info.model, UNet2DConditionModel)
             with (
                 ExitStack() as exit_stack,
-                unet_info as unet,
+                unet_info.model_on_device() as (model_state_dict, unet),
                 ModelPatcher.apply_freeu(unet, self.unet.freeu_config),
                 set_seamless(unet, self.unet.seamless_axes),  # FIXME
                 # Apply the LoRA after unet has been moved to its target device for faster patching.
-                ModelPatcher.apply_lora_unet(unet, _lora_loader()),
+                ModelPatcher.apply_lora_unet(
+                    unet,
+                    loras=_lora_loader(),
+                    model_state_dict=model_state_dict,
+                ),
             ):
                 assert isinstance(unet, UNet2DConditionModel)
                 latents = latents.to(device=unet.device, dtype=unet.dtype)

invokeai/backend/model_manager/load/load_base.py

@@ -4,10 +4,13 @@ Base class for model loading in InvokeAI.
 """
 
 from abc import ABC, abstractmethod
+from contextlib import contextmanager
 from dataclasses import dataclass
 from logging import Logger
 from pathlib import Path
-from typing import Any, Optional
+from typing import Any, Dict, Generator, Optional, Tuple
+
+import torch
 
 from invokeai.app.services.config import InvokeAIAppConfig
 from invokeai.backend.model_manager.config import (
@@ -21,7 +24,42 @@ from invokeai.backend.model_manager.load.model_cache.model_cache_base import Mod
 
 @dataclass
 class LoadedModel:
-    """Context manager object that mediates transfer from RAM<->VRAM."""
+    """
+    Context manager object that mediates transfer from RAM<->VRAM.
+
+    This is a context manager object that has two distinct APIs:
+
+    1. Older API (deprecated):
+    Use the LoadedModel object directly as a context manager.
+    It will move the model into VRAM (on CUDA devices), and
+    return the model in a form suitable for passing to torch.
+    Example:
+    ```
+    loaded_model_= loader.get_model_by_key('f13dd932', SubModelType('vae'))
+    with loaded_model as vae:
+      image = vae.decode(latents)[0]
+    ```
+
+    2. Newer API (recommended):
+    Call the LoadedModel's `model_on_device()` method in a
+    context. It returns a tuple consisting of a copy of
+    the model's state dict in CPU RAM followed by a copy
+    of the model in VRAM. The state dict is provided to allow
+    LoRAs and other model patchers to return the model to
+    its unpatched state without expensive copy and restore
+    operations.
+
+    Example:
+    ```
+    loaded_model_= loader.get_model_by_key('f13dd932', SubModelType('vae'))
+    with loaded_model.model_on_device() as (state_dict, vae):
+        image = vae.decode(latents)[0]
+    ```
+
+    The state_dict should be treated as a read-only object and
+    never modified. Also be aware that some loadable models do
+    not have a state_dict, in which case this value will be None.
+    """
 
     config: AnyModelConfig
     _locker: ModelLockerBase
@@ -35,6 +73,16 @@ class LoadedModel:
         """Context exit."""
         self._locker.unlock()
 
+    @contextmanager
+    def model_on_device(self) -> Generator[Tuple[Optional[Dict[str, torch.Tensor]], AnyModel], None, None]:
+        """Return a tuple consisting of the model's state dict (if it exists) and the locked model on execution device."""
+        locked_model = self._locker.lock()
+        try:
+            state_dict = self._locker.get_state_dict()
+            yield (state_dict, locked_model)
+        finally:
+            self._locker.unlock()
+
     @property
     def model(self) -> AnyModel:
         """Return the model without locking it."""

invokeai/backend/model_manager/load/model_cache/model_cache_base.py

@@ -30,6 +30,11 @@ class ModelLockerBase(ABC):
         """Unlock the contained model, and remove it from VRAM."""
         pass
 
+    @abstractmethod
+    def get_state_dict(self) -> Optional[Dict[str, torch.Tensor]]:
+        """Return the state dict (if any) for the cached model."""
+        pass
+
     @property
     @abstractmethod
     def model(self) -> AnyModel:
@@ -56,6 +61,11 @@ class CacheRecord(Generic[T]):
     and then injected into the model. When the model is finished, the VRAM
     copy of the state dict is deleted, and the RAM version is reinjected
     into the model.
+
+    The state_dict should be treated as a read-only attribute. Do not attempt
+    to patch or otherwise modify it. Instead, patch the copy of the state_dict
+    after it is loaded into the execution device (e.g. CUDA) using the `LoadedModel`
+    context manager call `model_on_device()`.
     """
 
     key: str

invokeai/backend/model_manager/load/model_cache/model_locker.py

@@ -2,6 +2,8 @@
 Base class and implementation of a class that moves models in and out of VRAM.
 """
 
+from typing import Dict, Optional
+
 import torch
 
 from invokeai.backend.model_manager import AnyModel
@@ -27,6 +29,10 @@ class ModelLocker(ModelLockerBase):
         """Return the model without moving it around."""
         return self._cache_entry.model
 
+    def get_state_dict(self) -> Optional[Dict[str, torch.Tensor]]:
+        """Return the state dict (if any) for the cached model."""
+        return self._cache_entry.state_dict
+
     def lock(self) -> AnyModel:
         """Move the model into the execution device (GPU) and lock it."""
         if not hasattr(self.model, "to"):
@@ -37,10 +43,8 @@ class ModelLocker(ModelLockerBase):
         try:
             if self._cache.lazy_offloading:
                 self._cache.offload_unlocked_models(self._cache_entry.size)
-
             self._cache.move_model_to_device(self._cache_entry, self._cache.execution_device)
             self._cache_entry.loaded = True
-
             self._cache.logger.debug(f"Locking {self._cache_entry.key} in {self._cache.execution_device}")
             self._cache.print_cuda_stats()
         except torch.cuda.OutOfMemoryError:

invokeai/backend/model_patcher.py

@@ -5,7 +5,7 @@ from __future__ import annotations
 
 import pickle
 from contextlib import contextmanager
-from typing import Any, Dict, Iterator, List, Optional, Tuple, Union
+from typing import Any, Dict, Generator, Iterator, List, Optional, Tuple, Union
 
 import numpy as np
 import torch
@@ -66,8 +66,14 @@ class ModelPatcher:
         cls,
         unet: UNet2DConditionModel,
         loras: Iterator[Tuple[LoRAModelRaw, float]],
+        model_state_dict: Optional[Dict[str, torch.Tensor]] = None,
     ) -> None:
-        with cls.apply_lora(unet, loras, "lora_unet_"):
+        with cls.apply_lora(
+            unet,
+            loras=loras,
+            prefix="lora_unet_",
+            model_state_dict=model_state_dict,
+        ):
             yield
 
     @classmethod
@@ -76,28 +82,9 @@ class ModelPatcher:
         cls,
         text_encoder: CLIPTextModel,
         loras: Iterator[Tuple[LoRAModelRaw, float]],
+        model_state_dict: Optional[Dict[str, torch.Tensor]] = None,
     ) -> None:
-        with cls.apply_lora(text_encoder, loras, "lora_te_"):
+        with cls.apply_lora(text_encoder, loras=loras, prefix="lora_te_", model_state_dict=model_state_dict):
-            yield
-
-    @classmethod
-    @contextmanager
-    def apply_sdxl_lora_text_encoder(
-        cls,
-        text_encoder: CLIPTextModel,
-        loras: List[Tuple[LoRAModelRaw, float]],
-    ) -> None:
-        with cls.apply_lora(text_encoder, loras, "lora_te1_"):
-            yield
-
-    @classmethod
-    @contextmanager
-    def apply_sdxl_lora_text_encoder2(
-        cls,
-        text_encoder: CLIPTextModel,
-        loras: List[Tuple[LoRAModelRaw, float]],
-    ) -> None:
-        with cls.apply_lora(text_encoder, loras, "lora_te2_"):
             yield
 
     @classmethod
@@ -107,7 +94,16 @@ class ModelPatcher:
         model: AnyModel,
         loras: Iterator[Tuple[LoRAModelRaw, float]],
         prefix: str,
-    ) -> None:
+        model_state_dict: Optional[Dict[str, torch.Tensor]] = None,
+    ) -> Generator[Any, None, None]:
+        """
+        Apply one or more LoRAs to a model.
+
+        :param model: The model to patch.
+        :param loras: An iterator that returns the LoRA to patch in and its patch weight.
+        :param prefix: A string prefix that precedes keys used in the LoRAs weight layers.
+        :model_state_dict: Read-only copy of the model's state dict in CPU, for unpatching purposes.
+        """
         original_weights = {}
         try:
             with torch.no_grad():
@@ -133,7 +129,10 @@ class ModelPatcher:
                         dtype = module.weight.dtype
 
                         if module_key not in original_weights:
-                            original_weights[module_key] = module.weight.detach().to(device="cpu", copy=True)
+                            if model_state_dict is not None:  # we were provided with the CPU copy of the state dict
+                                original_weights[module_key] = model_state_dict[module_key + ".weight"]
+                            else:
+                                original_weights[module_key] = module.weight.detach().to(device="cpu", copy=True)
 
                         layer_scale = layer.alpha / layer.rank if (layer.alpha and layer.rank) else 1.0