Add Self-Attention-Guided (SAG) Stable Diffusion pipeline (#2193)

* Add Stable Diffusion Sw/ elf-Attention Guidance

* Modify __init__.py

* Register attention storing processor

* Update pipeline_stable_diffusion_sag.py

* Editing default value

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Update dummy_torch_and_transformers_objects.py

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update pipeline_stable_diffusion_sag.py

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Create test_stable_diffusion_sag.py

* Create self_attention_guidance.py

* Update pipeline_stable_diffusion_sag.py

* Update test_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Rename self_attention_guidance.py to self_attention_guidance.mdx

* Update self_attention_guidance.mdx

* Update self_attention_guidance.mdx

* Update _toctree.yml

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Fixing order

* Update pipeline_stable_diffusion_sag.py

* fixing import order

* fix order

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* Naming change

* Noting pred_x0

* Adding some fast tests

* Update pipeline_stable_diffusion_sag.py

* Update test_stable_diffusion_sag.py

* Update test_stable_diffusion_sag.py

* Update test_stable_diffusion_sag.py

* Update docs/source/en/api/pipelines/stable_diffusion/self_attention_guidance.mdx

* implement gaussian_blur

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

* fix tests

* Update pipeline_stable_diffusion_sag.py

* Update pipeline_stable_diffusion_sag.py

---------

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Will Berman <wlbberman@gmail.com>

docs/source/en/_toctree.yml

@@ -153,6 +153,8 @@
         title: InstructPix2Pix
       - local: api/pipelines/stable_diffusion/pix2pix_zero
         title: Pix2Pix Zero
+      - local: api/pipelines/stable_diffusion/self_attention_guidance
+        title: Self-Attention Guidance
       title: Stable Diffusion
     - local: api/pipelines/stable_diffusion_2
       title: Stable Diffusion 2

docs/source/en/api/pipelines/stable_diffusion/self_attention_guidance.mdx

@@ -0,0 +1,64 @@
+<!--Copyright 2023 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+-->
+
+# Self-Attention Guidance (SAG)
+
+## Overview
+
+[Self-Attention Guidance](https://arxiv.org/abs/2210.00939) by Susung Hong et al.
+
+The abstract of the paper is the following:
+
+*Denoising diffusion models (DDMs) have been drawing much attention for their appreciable sample quality and diversity. Despite their remarkable performance, DDMs remain black boxes on which further study is necessary to take a profound step. Motivated by this, we delve into the design of conventional U-shaped diffusion models. More specifically, we investigate the self-attention modules within these models through carefully designed experiments and explore their characteristics. In addition, inspired by the studies that substantiate the effectiveness of the guidance schemes, we present plug-and-play diffusion guidance, namely Self-Attention Guidance (SAG), that can drastically boost the performance of existing diffusion models. Our method, SAG, extracts the intermediate attention map from a diffusion model at every iteration and selects tokens above a certain attention score for masking and blurring to obtain a partially blurred input. Subsequently, we measure the dissimilarity between the predicted noises obtained from feeding the blurred and original input to the diffusion model and leverage it as guidance. With this guidance, we observe apparent improvements in a wide range of diffusion models, e.g., ADM, IDDPM, and Stable Diffusion, and show that the results further improve by combining our method with the conventional guidance scheme. We provide extensive ablation studies to verify our choices.*
+
+Resources:
+
+* [Project Page](https://ku-cvlab.github.io/Self-Attention-Guidance).
+* [Paper](https://arxiv.org/abs/2210.00939).
+* [Original Code](https://github.com/KU-CVLAB/Self-Attention-Guidance).
+* [Demo](https://colab.research.google.com/github/SusungHong/Self-Attention-Guidance/blob/main/SAG_Stable.ipynb).
+
+
+## Available Pipelines:
+
+| Pipeline | Tasks | Demo
+|---|---|:---:|
+| [StableDiffusionSAGPipeline](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py) | *Text-to-Image Generation* | [Colab](https://colab.research.google.com/github/SusungHong/Self-Attention-Guidance/blob/main/SAG_Stable.ipynb) |
+
+## Usage example
+
+```python
+import torch
+from diffusers import StableDiffusionSAGPipeline
+from accelerate.utils import set_seed
+
+pipe = StableDiffusionSAGPipeline.from_pretrained("CompVis/stable-diffusion-v1-4", torch_dtype=torch.float16)
+pipe = pipe.to("cuda")
+
+seed = 8978
+prompt = "."
+guidance_scale = 7.5
+num_images_per_prompt = 1
+
+sag_scale = 1.0
+
+set_seed(seed)
+images = pipe(
+    prompt, num_images_per_prompt=num_images_per_prompt, guidance_scale=guidance_scale, sag_scale=sag_scale
+).images
+images[0].save("example.png")
+```
+
+## StableDiffusionSAGPipeline
+[[autodoc]] StableDiffusionSAGPipeline
+	- __call__
+	- all

src/diffusers/__init__.py

@@ -119,6 +119,7 @@ else:
         StableDiffusionPipeline,
         StableDiffusionPipelineSafe,
         StableDiffusionPix2PixZeroPipeline,
+        StableDiffusionSAGPipeline,
         StableDiffusionUpscalePipeline,
         StableUnCLIPImg2ImgPipeline,
         StableUnCLIPPipeline,

src/diffusers/pipelines/__init__.py

@@ -55,6 +55,7 @@ else:
         StableDiffusionLatentUpscalePipeline,
         StableDiffusionPipeline,
         StableDiffusionPix2PixZeroPipeline,
+        StableDiffusionSAGPipeline,
         StableDiffusionUpscalePipeline,
         StableUnCLIPImg2ImgPipeline,
         StableUnCLIPPipeline,

src/diffusers/pipelines/stable_diffusion/__init__.py

@@ -44,6 +44,7 @@ if is_transformers_available() and is_torch_available():
     from .pipeline_stable_diffusion_inpaint_legacy import StableDiffusionInpaintPipelineLegacy
     from .pipeline_stable_diffusion_instruct_pix2pix import StableDiffusionInstructPix2PixPipeline
     from .pipeline_stable_diffusion_latent_upscale import StableDiffusionLatentUpscalePipeline
+    from .pipeline_stable_diffusion_sag import StableDiffusionSAGPipeline
     from .pipeline_stable_diffusion_upscale import StableDiffusionUpscalePipeline
     from .pipeline_stable_unclip import StableUnCLIPPipeline
     from .pipeline_stable_unclip_img2img import StableUnCLIPImg2ImgPipeline

src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_sag.py

@@ -0,0 +1,763 @@
+# Copyright 2022 Susung Hong and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import inspect
+import math
+from typing import Any, Callable, Dict, List, Optional, Union
+
+import torch
+import torch.nn.functional as F
+from transformers import CLIPFeatureExtractor, CLIPTextModel, CLIPTokenizer
+
+from ...models import AutoencoderKL, UNet2DConditionModel
+from ...schedulers import KarrasDiffusionSchedulers
+from ...utils import is_accelerate_available, logging, randn_tensor, replace_example_docstring
+from ..pipeline_utils import DiffusionPipeline
+from . import StableDiffusionPipelineOutput
+from .safety_checker import StableDiffusionSafetyChecker
+
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> import torch
+        >>> from diffusers import StableDiffusionSAGPipeline
+
+        >>> pipe = StableDiffusionSAGPipeline.from_pretrained(
+        ...     "runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16
+        ... )
+        >>> pipe = pipe.to("cuda")
+
+        >>> prompt = "a photo of an astronaut riding a horse on mars"
+        >>> image = pipe(prompt, sag_scale=0.75).images[0]
+        ```
+"""
+
+
+# processes and stores attention probabilities
+class CrossAttnStoreProcessor:
+    def __init__(self):
+        self.attention_probs = None
+
+    def __call__(
+        self,
+        attn,
+        hidden_states,
+        encoder_hidden_states=None,
+        attention_mask=None,
+    ):
+        batch_size, sequence_length, _ = hidden_states.shape
+        attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length, batch_size)
+        query = attn.to_q(hidden_states)
+
+        if encoder_hidden_states is None:
+            encoder_hidden_states = hidden_states
+        elif attn.cross_attention_norm:
+            encoder_hidden_states = attn.norm_cross(encoder_hidden_states)
+
+        key = attn.to_k(encoder_hidden_states)
+        value = attn.to_v(encoder_hidden_states)
+
+        query = attn.head_to_batch_dim(query)
+        key = attn.head_to_batch_dim(key)
+        value = attn.head_to_batch_dim(value)
+
+        self.attention_probs = attn.get_attention_scores(query, key, attention_mask)
+        hidden_states = torch.bmm(self.attention_probs, value)
+        hidden_states = attn.batch_to_head_dim(hidden_states)
+
+        # linear proj
+        hidden_states = attn.to_out[0](hidden_states)
+        # dropout
+        hidden_states = attn.to_out[1](hidden_states)
+
+        return hidden_states
+
+
+# Modified to get self-attention guidance scale in this paper (https://arxiv.org/pdf/2210.00939.pdf) as an input
+class StableDiffusionSAGPipeline(DiffusionPipeline):
+    r"""
+    Pipeline for text-to-image generation using Stable Diffusion.
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
+    library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
+
+    Args:
+        vae ([`AutoencoderKL`]):
+            Variational Auto-Encoder (VAE) Model to encode and decode images to and from latent representations.
+        text_encoder ([`CLIPTextModel`]):
+            Frozen text-encoder. Stable Diffusion uses the text portion of
+            [CLIP](https://huggingface.co/docs/transformers/model_doc/clip#transformers.CLIPTextModel), specifically
+            the [clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14) variant.
+        tokenizer (`CLIPTokenizer`):
+            Tokenizer of class
+            [CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer).
+        unet ([`UNet2DConditionModel`]): Conditional U-Net architecture to denoise the encoded image latents.
+        scheduler ([`SchedulerMixin`]):
+            A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of
+            [`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`].
+        safety_checker ([`StableDiffusionSafetyChecker`]):
+            Classification module that estimates whether generated images could be considered offensive or harmful.
+            Please, refer to the [model card](https://huggingface.co/runwayml/stable-diffusion-v1-5) for details.
+        feature_extractor ([`CLIPFeatureExtractor`]):
+            Model that extracts features from generated images to be used as inputs for the `safety_checker`.
+    """
+    _optional_components = ["safety_checker", "feature_extractor"]
+
+    def __init__(
+        self,
+        vae: AutoencoderKL,
+        text_encoder: CLIPTextModel,
+        tokenizer: CLIPTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: KarrasDiffusionSchedulers,
+        safety_checker: StableDiffusionSafetyChecker,
+        feature_extractor: CLIPFeatureExtractor,
+        requires_safety_checker: bool = True,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            vae=vae,
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            safety_checker=safety_checker,
+            feature_extractor=feature_extractor,
+        )
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
+        self.register_to_config(requires_safety_checker=requires_safety_checker)
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.enable_vae_slicing
+    def enable_vae_slicing(self):
+        r"""
+        Enable sliced VAE decoding.
+
+        When this option is enabled, the VAE will split the input tensor in slices to compute decoding in several
+        steps. This is useful to save some memory and allow larger batch sizes.
+        """
+        self.vae.enable_slicing()
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.disable_vae_slicing
+    def disable_vae_slicing(self):
+        r"""
+        Disable sliced VAE decoding. If `enable_vae_slicing` was previously invoked, this method will go back to
+        computing decoding in one step.
+        """
+        self.vae.disable_slicing()
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.enable_sequential_cpu_offload
+    def enable_sequential_cpu_offload(self, gpu_id=0):
+        r"""
+        Offloads all models to CPU using accelerate, significantly reducing memory usage. When called, unet,
+        text_encoder, vae and safety checker have their state dicts saved to CPU and then are moved to a
+        `torch.device('meta') and loaded to GPU only when their specific submodule has its `forward` method called.
+        """
+        if is_accelerate_available():
+            from accelerate import cpu_offload
+        else:
+            raise ImportError("Please install accelerate via `pip install accelerate`")
+
+        device = torch.device(f"cuda:{gpu_id}")
+
+        for cpu_offloaded_model in [self.unet, self.text_encoder, self.vae]:
+            cpu_offload(cpu_offloaded_model, device)
+
+        if self.safety_checker is not None:
+            cpu_offload(self.safety_checker, execution_device=device, offload_buffers=True)
+
+    @property
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline._execution_device
+    def _execution_device(self):
+        r"""
+        Returns the device on which the pipeline's models will be executed. After calling
+        `pipeline.enable_sequential_cpu_offload()` the execution device can only be inferred from Accelerate's module
+        hooks.
+        """
+        if self.device != torch.device("meta") or not hasattr(self.unet, "_hf_hook"):
+            return self.device
+        for module in self.unet.modules():
+            if (
+                hasattr(module, "_hf_hook")
+                and hasattr(module._hf_hook, "execution_device")
+                and module._hf_hook.execution_device is not None
+            ):
+                return torch.device(module._hf_hook.execution_device)
+        return self.device
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline._encode_prompt
+    def _encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+        prompt_embeds: Optional[torch.FloatTensor] = None,
+        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+    ):
+        r"""
+        Encodes the prompt into text encoder hidden states.
+
+        Args:
+             prompt (`str` or `List[str]`, *optional*):
+                prompt to be encoded
+            device: (`torch.device`):
+                torch device
+            num_images_per_prompt (`int`):
+                number of images that should be generated per prompt
+            do_classifier_free_guidance (`bool`):
+                whether to use classifier free guidance or not
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. If not defined, one has to pass
+                `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
+                Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`).
+            prompt_embeds (`torch.FloatTensor`, *optional*):
+                Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
+                provided, text embeddings will be generated from `prompt` input argument.
+            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+                Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
+                weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
+                argument.
+        """
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        if prompt_embeds is None:
+            text_inputs = self.tokenizer(
+                prompt,
+                padding="max_length",
+                max_length=self.tokenizer.model_max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+            text_input_ids = text_inputs.input_ids
+            untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+            if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(
+                text_input_ids, untruncated_ids
+            ):
+                removed_text = self.tokenizer.batch_decode(
+                    untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1]
+                )
+                logger.warning(
+                    "The following part of your input was truncated because CLIP can only handle sequences up to"
+                    f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+                )
+
+            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+                attention_mask = text_inputs.attention_mask.to(device)
+            else:
+                attention_mask = None
+
+            prompt_embeds = self.text_encoder(
+                text_input_ids.to(device),
+                attention_mask=attention_mask,
+            )
+            prompt_embeds = prompt_embeds[0]
+
+        prompt_embeds = prompt_embeds.to(dtype=self.text_encoder.dtype, device=device)
+
+        bs_embed, seq_len, _ = prompt_embeds.shape
+        # duplicate text embeddings for each generation per prompt, using mps friendly method
+        prompt_embeds = prompt_embeds.repeat(1, num_images_per_prompt, 1)
+        prompt_embeds = prompt_embeds.view(bs_embed * num_images_per_prompt, seq_len, -1)
+
+        # get unconditional embeddings for classifier free guidance
+        if do_classifier_free_guidance and negative_prompt_embeds is None:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            max_length = prompt_embeds.shape[1]
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+
+            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+                attention_mask = uncond_input.attention_mask.to(device)
+            else:
+                attention_mask = None
+
+            negative_prompt_embeds = self.text_encoder(
+                uncond_input.input_ids.to(device),
+                attention_mask=attention_mask,
+            )
+            negative_prompt_embeds = negative_prompt_embeds[0]
+
+        if do_classifier_free_guidance:
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+            seq_len = negative_prompt_embeds.shape[1]
+
+            negative_prompt_embeds = negative_prompt_embeds.to(dtype=self.text_encoder.dtype, device=device)
+
+            negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt, 1)
+            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)
+
+            # For classifier free guidance, we need to do two forward passes.
+            # Here we concatenate the unconditional and text embeddings into a single batch
+            # to avoid doing two forward passes
+            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
+
+        return prompt_embeds
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.run_safety_checker
+    def run_safety_checker(self, image, device, dtype):
+        if self.safety_checker is not None:
+            safety_checker_input = self.feature_extractor(self.numpy_to_pil(image), return_tensors="pt").to(device)
+            image, has_nsfw_concept = self.safety_checker(
+                images=image, clip_input=safety_checker_input.pixel_values.to(dtype)
+            )
+        else:
+            has_nsfw_concept = None
+        return image, has_nsfw_concept
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.decode_latents
+    def decode_latents(self, latents):
+        latents = 1 / self.vae.config.scaling_factor * latents
+        image = self.vae.decode(latents).sample
+        image = (image / 2 + 0.5).clamp(0, 1)
+        # we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
+        image = image.cpu().permute(0, 2, 3, 1).float().numpy()
+        return image
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_extra_step_kwargs
+    def prepare_extra_step_kwargs(self, generator, eta):
+        # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
+        # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
+        # eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502
+        # and should be between [0, 1]
+
+        accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        extra_step_kwargs = {}
+        if accepts_eta:
+            extra_step_kwargs["eta"] = eta
+
+        # check if the scheduler accepts generator
+        accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        if accepts_generator:
+            extra_step_kwargs["generator"] = generator
+        return extra_step_kwargs
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.check_inputs
+    def check_inputs(
+        self,
+        prompt,
+        height,
+        width,
+        callback_steps,
+        negative_prompt=None,
+        prompt_embeds=None,
+        negative_prompt_embeds=None,
+    ):
+        if height % 8 != 0 or width % 8 != 0:
+            raise ValueError(f"`height` and `width` have to be divisible by 8 but are {height} and {width}.")
+
+        if (callback_steps is None) or (
+            callback_steps is not None and (not isinstance(callback_steps, int) or callback_steps <= 0)
+        ):
+            raise ValueError(
+                f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
+                f" {type(callback_steps)}."
+            )
+
+        if prompt is not None and prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
+                " only forward one of the two."
+            )
+        elif prompt is None and prompt_embeds is None:
+            raise ValueError(
+                "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
+            )
+        elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        if negative_prompt is not None and negative_prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
+                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
+            )
+
+        if prompt_embeds is not None and negative_prompt_embeds is not None:
+            if prompt_embeds.shape != negative_prompt_embeds.shape:
+                raise ValueError(
+                    "`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
+                    f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
+                    f" {negative_prompt_embeds.shape}."
+                )
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_latents
+    def prepare_latents(self, batch_size, num_channels_latents, height, width, dtype, device, generator, latents=None):
+        shape = (batch_size, num_channels_latents, height // self.vae_scale_factor, width // self.vae_scale_factor)
+        if isinstance(generator, list) and len(generator) != batch_size:
+            raise ValueError(
+                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
+                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+            )
+
+        if latents is None:
+            latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        else:
+            latents = latents.to(device)
+
+        # scale the initial noise by the standard deviation required by the scheduler
+        latents = latents * self.scheduler.init_noise_sigma
+        return latents
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]] = None,
+        height: Optional[int] = None,
+        width: Optional[int] = None,
+        num_inference_steps: int = 50,
+        guidance_scale: float = 7.5,
+        sag_scale: float = 0.75,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        num_images_per_prompt: Optional[int] = 1,
+        eta: float = 0.0,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.FloatTensor] = None,
+        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        output_type: Optional[str] = "pil",
+        return_dict: bool = True,
+        callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
+        callback_steps: Optional[int] = 1,
+        cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+    ):
+        r"""
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
+                instead.
+            height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
+                The width in pixels of the generated image.
+            num_inference_steps (`int`, *optional*, defaults to 50):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            guidance_scale (`float`, *optional*, defaults to 7.5):
+                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
+                `guidance_scale` is defined as `w` of equation 2. of [Imagen
+                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
+                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
+                usually at the expense of lower image quality.
+            sag_scale (`float`, *optional*, defaults to 0.75):
+                SAG scale as defined in [Improving Sample Quality of Diffusion Models Using Self-Attention Guidance]
+                (https://arxiv.org/abs/2210.00939). `sag_scale` is defined as `s_s` of equation (24) of SAG paper:
+                https://arxiv.org/pdf/2210.00939.pdf. Typically chosen between [0, 1.0] for better quality.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. If not defined, one has to pass
+                `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
+                Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`).
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            eta (`float`, *optional*, defaults to 0.0):
+                Corresponds to parameter eta (η) in the DDIM paper: https://arxiv.org/abs/2010.02502. Only applies to
+                [`schedulers.DDIMScheduler`], will be ignored for others.
+            generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
+                One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
+                to make generation deterministic.
+            latents (`torch.FloatTensor`, *optional*):
+                Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
+                generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
+                tensor will ge generated by sampling using the supplied random `generator`.
+            prompt_embeds (`torch.FloatTensor`, *optional*):
+                Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
+                provided, text embeddings will be generated from `prompt` input argument.
+            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+                Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
+                weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
+                argument.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generate image. Choose between
+                [PIL](https://pillow.readthedocs.io/en/stable/): `PIL.Image.Image` or `np.array`.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] instead of a
+                plain tuple.
+            callback (`Callable`, *optional*):
+                A function that will be called every `callback_steps` steps during inference. The function will be
+                called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`.
+            callback_steps (`int`, *optional*, defaults to 1):
+                The frequency at which the `callback` function will be called. If not specified, the callback will be
+                called at every step.
+            cross_attention_kwargs (`dict`, *optional*):
+                A kwargs dictionary that if specified is passed along to the `AttnProcessor` as defined under
+                `self.processor` in
+                [diffusers.cross_attention](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/cross_attention.py).
+
+        Examples:
+
+        Returns:
+            [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] or `tuple`:
+            [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] if `return_dict` is True, otherwise a `tuple.
+            When returning a tuple, the first element is a list with the generated images, and the second element is a
+            list of `bool`s denoting whether the corresponding generated image likely represents "not-safe-for-work"
+            (nsfw) content, according to the `safety_checker`.
+        """
+        # 0. Default height and width to unet
+        height = height or self.unet.config.sample_size * self.vae_scale_factor
+        width = width or self.unet.config.sample_size * self.vae_scale_factor
+
+        # 1. Check inputs. Raise error if not correct
+        self.check_inputs(
+            prompt, height, width, callback_steps, negative_prompt, prompt_embeds, negative_prompt_embeds
+        )
+
+        # 2. Define call parameters
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        device = self._execution_device
+        # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
+        # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
+        # corresponds to doing no classifier free guidance.
+        do_classifier_free_guidance = guidance_scale > 1.0
+        # and `sag_scale` is` `s` of equation (15)
+        # of the self-attentnion guidance paper: https://arxiv.org/pdf/2210.00939.pdf
+        # `sag_scale = 0` means no self-attention guidance
+        do_self_attention_guidance = sag_scale > 0.0
+
+        # 3. Encode input prompt
+        prompt_embeds = self._encode_prompt(
+            prompt,
+            device,
+            num_images_per_prompt,
+            do_classifier_free_guidance,
+            negative_prompt,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
+        )
+
+        # 4. Prepare timesteps
+        self.scheduler.set_timesteps(num_inference_steps, device=device)
+        timesteps = self.scheduler.timesteps
+
+        # 5. Prepare latent variables
+        num_channels_latents = self.unet.in_channels
+        latents = self.prepare_latents(
+            batch_size * num_images_per_prompt,
+            num_channels_latents,
+            height,
+            width,
+            prompt_embeds.dtype,
+            device,
+            generator,
+            latents,
+        )
+
+        # 6. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
+        extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)
+
+        # 7. Denoising loop
+        store_processor = CrossAttnStoreProcessor()
+        self.unet.mid_block.attentions[0].transformer_blocks[0].attn1.processor = store_processor
+        num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order
+        with self.progress_bar(total=num_inference_steps) as progress_bar:
+            for i, t in enumerate(timesteps):
+                # expand the latents if we are doing classifier free guidance
+                latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
+                latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
+
+                # predict the noise residual
+                noise_pred = self.unet(
+                    latent_model_input,
+                    t,
+                    encoder_hidden_states=prompt_embeds,
+                    cross_attention_kwargs=cross_attention_kwargs,
+                ).sample
+
+                # perform guidance
+                if do_classifier_free_guidance:
+                    noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
+                    noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
+
+                # perform self-attention guidance with the stored self-attentnion map
+                if do_self_attention_guidance:
+                    # classifier-free guidance produces two chunks of attention map
+                    # and we only use unconditional one according to equation (24)
+                    # in https://arxiv.org/pdf/2210.00939.pdf
+                    if do_classifier_free_guidance:
+                        # DDIM-like prediction of x0
+                        pred_x0 = self.pred_x0(latents, noise_pred_uncond, t)
+                        # get the stored attention maps
+                        uncond_attn, cond_attn = store_processor.attention_probs.chunk(2)
+                        # self-attention-based degrading of latents
+                        degraded_latents = self.sag_masking(
+                            pred_x0, uncond_attn, t, self.pred_epsilon(latents, noise_pred_uncond, t)
+                        )
+                        uncond_emb, _ = prompt_embeds.chunk(2)
+                        # forward and give guidance
+                        degraded_pred = self.unet(degraded_latents, t, encoder_hidden_states=uncond_emb).sample
+                        noise_pred += sag_scale * (noise_pred_uncond - degraded_pred)
+                    else:
+                        # DDIM-like prediction of x0
+                        pred_x0 = self.pred_x0(latents, noise_pred, t)
+                        # get the stored attention maps
+                        cond_attn = store_processor.attention_probs
+                        # self-attention-based degrading of latents
+                        degraded_latents = self.sag_masking(
+                            pred_x0, cond_attn, t, self.pred_epsilon(latents, noise_pred, t)
+                        )
+                        # forward and give guidance
+                        degraded_pred = self.unet(degraded_latents, t, encoder_hidden_states=prompt_embeds).sample
+                        noise_pred += sag_scale * (noise_pred - degraded_pred)
+
+                # compute the previous noisy sample x_t -> x_t-1
+                latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs).prev_sample
+
+                # call the callback, if provided
+                if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
+                    progress_bar.update()
+                    if callback is not None and i % callback_steps == 0:
+                        callback(i, t, latents)
+
+        # 8. Post-processing
+        image = self.decode_latents(latents)
+
+        # 9. Run safety checker
+        image, has_nsfw_concept = self.run_safety_checker(image, device, prompt_embeds.dtype)
+
+        # 10. Convert to PIL
+        if output_type == "pil":
+            image = self.numpy_to_pil(image)
+
+        if not return_dict:
+            return (image, has_nsfw_concept)
+
+        return StableDiffusionPipelineOutput(images=image, nsfw_content_detected=has_nsfw_concept)
+
+    def sag_masking(self, original_latents, attn_map, t, eps):
+        # Same masking process as in SAG paper: https://arxiv.org/pdf/2210.00939.pdf
+        bh, hw1, hw2 = attn_map.shape
+        b, latent_channel, latent_h, latent_w = original_latents.shape
+        h = self.unet.attention_head_dim
+        if isinstance(h, list):
+            h = h[-1]
+        map_size = math.isqrt(hw1)
+
+        # Produce attention mask
+        attn_map = attn_map.reshape(b, h, hw1, hw2)
+        attn_mask = attn_map.mean(1, keepdim=False).sum(1, keepdim=False) > 1.0
+        attn_mask = (
+            attn_mask.reshape(b, map_size, map_size).unsqueeze(1).repeat(1, latent_channel, 1, 1).type(attn_map.dtype)
+        )
+        attn_mask = F.interpolate(attn_mask, (latent_h, latent_w))
+
+        # Blur according to the self-attention mask
+        degraded_latents = gaussian_blur_2d(original_latents, kernel_size=9, sigma=1.0)
+        degraded_latents = degraded_latents * attn_mask + original_latents * (1 - attn_mask)
+
+        # Noise it again to match the noise level
+        degraded_latents = self.scheduler.add_noise(degraded_latents, noise=eps, timesteps=t)
+
+        return degraded_latents
+
+    # Modified from diffusers.schedulers.scheduling_ddim.DDIMScheduler.step
+    # Note: there are some schedulers that clip or do not return x_0 (PNDMScheduler, DDIMScheduler, etc.)
+    def pred_x0(self, sample, model_output, timestep):
+        alpha_prod_t = self.scheduler.alphas_cumprod[timestep]
+
+        beta_prod_t = 1 - alpha_prod_t
+        if self.scheduler.config.prediction_type == "epsilon":
+            pred_original_sample = (sample - beta_prod_t ** (0.5) * model_output) / alpha_prod_t ** (0.5)
+        elif self.scheduler.config.prediction_type == "sample":
+            pred_original_sample = model_output
+        elif self.scheduler.config.prediction_type == "v_prediction":
+            pred_original_sample = (alpha_prod_t**0.5) * sample - (beta_prod_t**0.5) * model_output
+            # predict V
+            model_output = (alpha_prod_t**0.5) * model_output + (beta_prod_t**0.5) * sample
+        else:
+            raise ValueError(
+                f"prediction_type given as {self.scheduler.config.prediction_type} must be one of `epsilon`, `sample`,"
+                " or `v_prediction`"
+            )
+
+        return pred_original_sample
+
+    def pred_epsilon(self, sample, model_output, timestep):
+        alpha_prod_t = self.scheduler.alphas_cumprod[timestep]
+
+        beta_prod_t = 1 - alpha_prod_t
+        if self.scheduler.config.prediction_type == "epsilon":
+            pred_eps = model_output
+        elif self.scheduler.config.prediction_type == "sample":
+            pred_eps = (sample - (alpha_prod_t**0.5) * model_output) / (beta_prod_t**0.5)
+        elif self.scheduler.config.prediction_type == "v_prediction":
+            pred_eps = (beta_prod_t**0.5) * sample + (alpha_prod_t**0.5) * model_output
+        else:
+            raise ValueError(
+                f"prediction_type given as {self.scheduler.config.prediction_type} must be one of `epsilon`, `sample`,"
+                " or `v_prediction`"
+            )
+
+        return pred_eps
+
+
+# Gaussian blur
+def gaussian_blur_2d(img, kernel_size, sigma):
+    ksize_half = (kernel_size - 1) * 0.5
+
+    x = torch.linspace(-ksize_half, ksize_half, steps=kernel_size)
+
+    pdf = torch.exp(-0.5 * (x / sigma).pow(2))
+
+    x_kernel = pdf / pdf.sum()
+    x_kernel = x_kernel.to(device=img.device, dtype=img.dtype)
+
+    kernel2d = torch.mm(x_kernel[:, None], x_kernel[None, :])
+    kernel2d = kernel2d.expand(img.shape[-3], 1, kernel2d.shape[0], kernel2d.shape[1])
+
+    padding = [kernel_size // 2, kernel_size // 2, kernel_size // 2, kernel_size // 2]
+
+    img = F.pad(img, padding, mode="reflect")
+    img = F.conv2d(img, kernel2d, groups=img.shape[-3])
+
+    return img

src/diffusers/utils/dummy_torch_and_transformers_objects.py

@@ -212,6 +212,21 @@ class StableDiffusionPipelineSafe(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class StableDiffusionSAGPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class StableDiffusionPix2PixZeroPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 

tests/pipelines/stable_diffusion/test_stable_diffusion_sag.py

@@ -0,0 +1,159 @@
+# coding=utf-8
+# Copyright 2022 HuggingFace Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import gc
+import unittest
+
+import numpy as np
+import torch
+from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
+
+from diffusers import (
+    AutoencoderKL,
+    DDIMScheduler,
+    StableDiffusionSAGPipeline,
+    UNet2DConditionModel,
+)
+from diffusers.utils import slow, torch_device
+from diffusers.utils.testing_utils import require_torch_gpu
+
+from ...test_pipelines_common import PipelineTesterMixin
+
+
+torch.backends.cuda.matmul.allow_tf32 = False
+
+
+class StableDiffusionSAGPipelineFastTests(PipelineTesterMixin, unittest.TestCase):
+    pipeline_class = StableDiffusionSAGPipeline
+    test_cpu_offload = False
+
+    def get_dummy_components(self):
+        torch.manual_seed(0)
+        unet = UNet2DConditionModel(
+            block_out_channels=(32, 64),
+            layers_per_block=2,
+            sample_size=32,
+            in_channels=4,
+            out_channels=4,
+            down_block_types=("DownBlock2D", "CrossAttnDownBlock2D"),
+            up_block_types=("CrossAttnUpBlock2D", "UpBlock2D"),
+            cross_attention_dim=32,
+        )
+        scheduler = DDIMScheduler(
+            beta_start=0.00085,
+            beta_end=0.012,
+            beta_schedule="scaled_linear",
+            clip_sample=False,
+            set_alpha_to_one=False,
+        )
+        torch.manual_seed(0)
+        vae = AutoencoderKL(
+            block_out_channels=[32, 64],
+            in_channels=3,
+            out_channels=3,
+            down_block_types=["DownEncoderBlock2D", "DownEncoderBlock2D"],
+            up_block_types=["UpDecoderBlock2D", "UpDecoderBlock2D"],
+            latent_channels=4,
+        )
+        torch.manual_seed(0)
+        text_encoder_config = CLIPTextConfig(
+            bos_token_id=0,
+            eos_token_id=2,
+            hidden_size=32,
+            intermediate_size=37,
+            layer_norm_eps=1e-05,
+            num_attention_heads=4,
+            num_hidden_layers=5,
+            pad_token_id=1,
+            vocab_size=1000,
+        )
+        text_encoder = CLIPTextModel(text_encoder_config)
+        tokenizer = CLIPTokenizer.from_pretrained("hf-internal-testing/tiny-random-clip")
+
+        components = {
+            "unet": unet,
+            "scheduler": scheduler,
+            "vae": vae,
+            "text_encoder": text_encoder,
+            "tokenizer": tokenizer,
+            "safety_checker": None,
+            "feature_extractor": None,
+        }
+        return components
+
+    def get_dummy_inputs(self, device, seed=0):
+        if str(device).startswith("mps"):
+            generator = torch.manual_seed(seed)
+        else:
+            generator = torch.Generator(device=device).manual_seed(seed)
+        inputs = {
+            "prompt": ".",
+            "generator": generator,
+            "num_inference_steps": 2,
+            "guidance_scale": 1.0,
+            "sag_scale": 1.0,
+            "output_type": "numpy",
+        }
+        return inputs
+
+
+@slow
+@require_torch_gpu
+class StableDiffusionPipelineIntegrationTests(unittest.TestCase):
+    def tearDown(self):
+        # clean up the VRAM after each test
+        super().tearDown()
+        gc.collect()
+        torch.cuda.empty_cache()
+
+    def test_stable_diffusion_1(self):
+        sag_pipe = StableDiffusionSAGPipeline.from_pretrained("CompVis/stable-diffusion-v1-4")
+        sag_pipe = sag_pipe.to(torch_device)
+        sag_pipe.set_progress_bar_config(disable=None)
+
+        prompt = "."
+        generator = torch.manual_seed(0)
+        output = sag_pipe(
+            [prompt], generator=generator, guidance_scale=7.5, sag_scale=1.0, num_inference_steps=20, output_type="np"
+        )
+
+        image = output.images
+
+        image_slice = image[0, -3:, -3:, -1]
+
+        assert image.shape == (1, 512, 512, 3)
+        expected_slice = np.array([0.1568, 0.1738, 0.1695, 0.1693, 0.1507, 0.1705, 0.1547, 0.1751, 0.1949])
+
+        assert np.abs(image_slice.flatten() - expected_slice).max() < 5e-4
+
+    def test_stable_diffusion_2(self):
+        sag_pipe = StableDiffusionSAGPipeline.from_pretrained("stabilityai/stable-diffusion-2-1-base")
+        sag_pipe = sag_pipe.to(torch_device)
+        sag_pipe.set_progress_bar_config(disable=None)
+
+        prompt = "."
+        generator = torch.manual_seed(0)
+        output = sag_pipe(
+            [prompt], generator=generator, guidance_scale=7.5, sag_scale=1.0, num_inference_steps=20, output_type="np"
+        )
+
+        image = output.images
+
+        image_slice = image[0, -3:, -3:, -1]
+
+        assert image.shape == (1, 512, 512, 3)
+        expected_slice = np.array([0.3459, 0.2876, 0.2537, 0.3002, 0.2671, 0.2160, 0.3026, 0.2262, 0.2371])
+
+        assert np.abs(image_slice.flatten() - expected_slice).max() < 5e-5