[Versatile Diffusion] Add versatile diffusion model (#1283)

* up

* convert dual unet

* revert dual attn

* adapt for vd-official

* test the full pipeline

* mixed inference

* mixed inference for text2img

* add image prompting

* fix clip norm

* split text2img and img2img

* fix format

* refactor text2img

* mega pipeline

* add optimus

* refactor image var

* wip text_unet

* text unet end to end

* update tests

* reshape

* fix image to text

* add some first docs

* dual guided pipeline

* fix token ratio

* propose change

* dual transformer as a native module

* DualTransformer(nn.Module)

* DualTransformer(nn.Module)

* correct unconditional image

* save-load with mega pipeline

* remove image to text

* up

* uP

* fix

* up

* final fix

* remove_unused_weights

* test updates

* save progress

* uP

* fix dual prompts

* some fixes

* finish

* style

* finish renaming

* up

* fix

* fix

* fix

* finish
Co-authored-by: anton-l <anton@huggingface.co>

docs/source/_toctree.yml

@@ -112,6 +112,8 @@
       title: "Stochastic Karras VE"
     - local: api/pipelines/dance_diffusion
       title: "Dance Diffusion"
+    - local: api/pipelines/versatile_diffusion
+      title: "Versatile Diffusion"
     - local: api/pipelines/vq_diffusion
       title: "VQ Diffusion"
     - local: api/pipelines/repaint

docs/source/api/pipelines/overview.mdx

@@ -60,6 +60,9 @@ available a colab notebook to directly try them out.
 | [stable_diffusion](./api/pipelines/stable_diffusion) | [**Stable Diffusion**](https://stability.ai/blog/stable-diffusion-public-release) | Text-Guided Image Inpainting | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/in_painting_with_stable_diffusion_using_diffusers.ipynb)
 | [stable_diffusion_safe](./api/pipelines/stable_diffusion_safe) | [**Safe Stable Diffusion**](https://arxiv.org/abs/2211.05105) | Text-Guided Generation | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ml-research/safe-latent-diffusion/blob/main/examples/Safe%20Latent%20Diffusion.ipynb)
 | [stochastic_karras_ve](./api/pipelines/stochastic_karras_ve) | [**Elucidating the Design Space of Diffusion-Based Generative Models**](https://arxiv.org/abs/2206.00364) | Unconditional Image Generation |
+| [versatile_diffusion](./api/pipelines/versatile_diffusion) | [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) | Text-to-Image Generation | 
+| [versatile_diffusion](./api/pipelines/versatile_diffusion) | [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) | Image Variations Generation | 
+| [versatile_diffusion](./api/pipelines/versatile_diffusion) | [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) | Dual Image and Text Guided Generation | 
 | [vq_diffusion](./api/pipelines/vq_diffusion) | [Vector Quantized Diffusion Model for Text-to-Image Synthesis](https://arxiv.org/abs/2111.14822) | Text-to-Image Generation | 
 
 

docs/source/api/pipelines/versatile_diffusion.mdx

+<!--Copyright 2022 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.
+-->
+
+# VersatileDiffusion
+
+VersatileDiffusion was proposed in [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) by Xingqian Xu, Zhangyang Wang, Eric Zhang, Kai Wang, Humphrey Shi .
+
+The abstract of the paper is the following:
+
+*The recent advances in diffusion models have set an impressive milestone in many generation tasks. Trending works such as DALL-E2, Imagen, and Stable Diffusion have attracted great interest in academia and industry. Despite the rapid landscape changes, recent new approaches focus on extensions and performance rather than capacity, thus requiring separate models for separate tasks. In this work, we expand the existing single-flow diffusion pipeline into a multi-flow network, dubbed Versatile Diffusion (VD), that handles text-to-image, image-to-text, image-variation, and text-variation in one unified model. Moreover, we generalize VD to a unified multi-flow multimodal diffusion framework with grouped layers, swappable streams, and other propositions that can process modalities beyond images and text. Through our experiments, we demonstrate that VD and its underlying framework have the following merits: a) VD handles all subtasks with competitive quality; b) VD initiates novel extensions and applications such as disentanglement of style and semantic, image-text dual-guided generation, etc.; c) Through these experiments and applications, VD provides more semantic insights of the generated outputs.*
+
+## Tips
+
+- VersatileDiffusion is conceptually very similar as [Stable Diffusion](./api/pipelines/stable_diffusion),  but instead of providing just a image data stream conditioned on text, VersatileDiffusion provides both a image and text data stream and can be conditioned on both text and image.
+
+### *Run VersatileDiffusion*
+
+You can both load the memory intensive "all-in-one" [`VersatileDiffusionPipeline`] that can run all tasks 
+with the same class as shown in [`VersatileDiffusionPipeline.text_to_image`], [`VersatileDiffusionPipeline.image_variation`], and [`VersatileDiffusionPipeline.dual_guided`]
+
+**or**
+
+You can run the individual pipelines which are much more memory efficient:
+
+- *Text-to-Image*: [`VersatileDiffusionTextToImagePipeline.__call__`]
+- *Image Variation*: [`VersatileDiffusionImageVariationPipeline.__call__`]
+- *Dual Text and Image Guided Generation*: [`VersatileDiffusionDualGuidedPipeline.__call__`]
+
+### *How to load and use different schedulers.*
+
+The versatile diffusion pipelines uses [`DDIMScheduler`] scheduler by default. But `diffusers` provides many other schedulers that can be used with the alt diffusion pipeline such as [`PNDMScheduler`], [`LMSDiscreteScheduler`], [`EulerDiscreteScheduler`], [`EulerAncestralDiscreteScheduler`] etc.
+To use a different scheduler, you can either change it via the [`ConfigMixin.from_config`] method or pass the `scheduler` argument to the `from_pretrained` method of the pipeline. For example, to use the [`EulerDiscreteScheduler`], you can do the following:
+
+```python
+>>> from diffusers import VersatileDiffusionPipeline, EulerDiscreteScheduler
+
+>>> pipeline = VersatileDiffusionPipeline.from_pretrained("shi-labs/versatile-diffusion")
+>>> pipeline.scheduler = EulerDiscreteScheduler.from_config(pipeline.scheduler.config)
+
+>>> # or
+>>> euler_scheduler = EulerDiscreteScheduler.from_pretrained("shi-labs/versatile-diffusion", subfolder="scheduler")
+>>> pipeline = VersatileDiffusionPipeline.from_pretrained("shi-labs/versatile-diffusion", scheduler=euler_scheduler)
+```
+
+## VersatileDiffusionPipeline
+[[autodoc]] VersatileDiffusionPipeline
+
+## VersatileDiffusionTextToImagePipeline
+[[autodoc]] VersatileDiffusionTextToImagePipeline
+	- __call__
+	- enable_attention_slicing
+	- disable_attention_slicing
+
+## VersatileDiffusionImageVariationPipeline
+[[autodoc]] VersatileDiffusionImageVariationPipeline
+	- __call__
+	- enable_attention_slicing
+	- disable_attention_slicing
+
+## VersatileDiffusionDualGuidedPipeline
+[[autodoc]] VersatileDiffusionDualGuidedPipeline
+	- __call__
+	- enable_attention_slicing
+	- disable_attention_slicing

docs/source/index.mdx

@@ -50,6 +50,9 @@ available a colab notebook to directly try them out.
 | [stable_diffusion](./api/pipelines/stable_diffusion) | [**Stable Diffusion**](https://stability.ai/blog/stable-diffusion-public-release) | Text-Guided Image Inpainting | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/in_painting_with_stable_diffusion_using_diffusers.ipynb)
 | [stable_diffusion_safe](./api/pipelines/stable_diffusion_safe) | [**Safe Stable Diffusion**](https://arxiv.org/abs/2211.05105) | Text-Guided Generation | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ml-research/safe-latent-diffusion/blob/main/examples/Safe%20Latent%20Diffusion.ipynb)
 | [stochastic_karras_ve](./api/pipelines/stochastic_karras_ve) | [**Elucidating the Design Space of Diffusion-Based Generative Models**](https://arxiv.org/abs/2206.00364) | Unconditional Image Generation | 
+| [versatile_diffusion](./api/pipelines/versatile_diffusion) | [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) | Text-to-Image Generation | 
+| [versatile_diffusion](./api/pipelines/versatile_diffusion) | [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) | Image Variations Generation | 
+| [versatile_diffusion](./api/pipelines/versatile_diffusion) | [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) | Dual Image and Text Guided Generation | 
 | [vq_diffusion](./api/pipelines/vq_diffusion) | [Vector Quantized Diffusion Model for Text-to-Image Synthesis](https://arxiv.org/abs/2111.14822) | Text-to-Image Generation | 
 
 **Note**: Pipelines are simple examples of how to play around with the diffusion systems as described in the corresponding papers. 

src/diffusers/__init__.py

@@ -75,6 +75,10 @@ if is_torch_available() and is_transformers_available():
         StableDiffusionInpaintPipelineLegacy,
         StableDiffusionPipeline,
         StableDiffusionPipelineSafe,
+        VersatileDiffusionDualGuidedPipeline,
+        VersatileDiffusionImageVariationPipeline,
+        VersatileDiffusionPipeline,
+        VersatileDiffusionTextToImagePipeline,
         VQDiffusionPipeline,
     )
 else:

src/diffusers/models/attention.py

@@ -699,3 +699,129 @@ class AdaLayerNorm(nn.Module):
         scale, shift = torch.chunk(emb, 2)
         x = self.norm(x) * (1 + scale) + shift
         return x
+
+
+class DualTransformer2DModel(nn.Module):
+    """
+    Dual transformer wrapper that combines two `Transformer2DModel`s for mixed inference.
+
+    Parameters:
+        num_attention_heads (`int`, *optional*, defaults to 16): The number of heads to use for multi-head attention.
+        attention_head_dim (`int`, *optional*, defaults to 88): The number of channels in each head.
+        in_channels (`int`, *optional*):
+            Pass if the input is continuous. The number of channels in the input and output.
+        num_layers (`int`, *optional*, defaults to 1): The number of layers of Transformer blocks to use.
+        dropout (`float`, *optional*, defaults to 0.1): The dropout probability to use.
+        cross_attention_dim (`int`, *optional*): The number of context dimensions to use.
+        sample_size (`int`, *optional*): Pass if the input is discrete. The width of the latent images.
+            Note that this is fixed at training time as it is used for learning a number of position embeddings. See
+            `ImagePositionalEmbeddings`.
+        num_vector_embeds (`int`, *optional*):
+            Pass if the input is discrete. The number of classes of the vector embeddings of the latent pixels.
+            Includes the class for the masked latent pixel.
+        activation_fn (`str`, *optional*, defaults to `"geglu"`): Activation function to be used in feed-forward.
+        num_embeds_ada_norm ( `int`, *optional*): Pass if at least one of the norm_layers is `AdaLayerNorm`.
+            The number of diffusion steps used during training. Note that this is fixed at training time as it is used
+            to learn a number of embeddings that are added to the hidden states. During inference, you can denoise for
+            up to but not more than steps than `num_embeds_ada_norm`.
+        attention_bias (`bool`, *optional*):
+            Configure if the TransformerBlocks' attention should contain a bias parameter.
+    """
+
+    def __init__(
+        self,
+        num_attention_heads: int = 16,
+        attention_head_dim: int = 88,
+        in_channels: Optional[int] = None,
+        num_layers: int = 1,
+        dropout: float = 0.0,
+        norm_num_groups: int = 32,
+        cross_attention_dim: Optional[int] = None,
+        attention_bias: bool = False,
+        sample_size: Optional[int] = None,
+        num_vector_embeds: Optional[int] = None,
+        activation_fn: str = "geglu",
+        num_embeds_ada_norm: Optional[int] = None,
+    ):
+        super().__init__()
+        self.transformers = nn.ModuleList(
+            [
+                Transformer2DModel(
+                    num_attention_heads=num_attention_heads,
+                    attention_head_dim=attention_head_dim,
+                    in_channels=in_channels,
+                    num_layers=num_layers,
+                    dropout=dropout,
+                    norm_num_groups=norm_num_groups,
+                    cross_attention_dim=cross_attention_dim,
+                    attention_bias=attention_bias,
+                    sample_size=sample_size,
+                    num_vector_embeds=num_vector_embeds,
+                    activation_fn=activation_fn,
+                    num_embeds_ada_norm=num_embeds_ada_norm,
+                )
+                for _ in range(2)
+            ]
+        )
+
+        # Variables that can be set by a pipeline:
+
+        # The ratio of transformer1 to transformer2's output states to be combined during inference
+        self.mix_ratio = 0.5
+
+        # The shape of `encoder_hidden_states` is expected to be
+        # `(batch_size, condition_lengths[0]+condition_lengths[1], num_features)`
+        self.condition_lengths = [77, 257]
+
+        # Which transformer to use to encode which condition.
+        # E.g. `(1, 0)` means that we'll use `transformers[1](conditions[0])` and `transformers[0](conditions[1])`
+        self.transformer_index_for_condition = [1, 0]
+
+    def forward(self, hidden_states, encoder_hidden_states, timestep=None, return_dict: bool = True):
+        """
+        Args:
+            hidden_states ( When discrete, `torch.LongTensor` of shape `(batch size, num latent pixels)`.
+                When continuous, `torch.FloatTensor` of shape `(batch size, channel, height, width)`): Input
+                hidden_states
+            encoder_hidden_states ( `torch.LongTensor` of shape `(batch size, context dim)`, *optional*):
+                Conditional embeddings for cross attention layer. If not given, cross-attention defaults to
+                self-attention.
+            timestep ( `torch.long`, *optional*):
+                Optional timestep to be applied as an embedding in AdaLayerNorm's. Used to indicate denoising step.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`models.unet_2d_condition.UNet2DConditionOutput`] instead of a plain tuple.
+
+        Returns:
+            [`~models.attention.Transformer2DModelOutput`] or `tuple`: [`~models.attention.Transformer2DModelOutput`]
+            if `return_dict` is True, otherwise a `tuple`. When returning a tuple, the first element is the sample
+            tensor.
+        """
+        input_states = hidden_states
+
+        encoded_states = []
+        tokens_start = 0
+        for i in range(2):
+            # for each of the two transformers, pass the corresponding condition tokens
+            condition_state = encoder_hidden_states[:, tokens_start : tokens_start + self.condition_lengths[i]]
+            transformer_index = self.transformer_index_for_condition[i]
+            encoded_state = self.transformers[transformer_index](input_states, condition_state, timestep, return_dict)[
+                0
+            ]
+            encoded_states.append(encoded_state - input_states)
+            tokens_start += self.condition_lengths[i]
+
+        output_states = encoded_states[0] * self.mix_ratio + encoded_states[1] * (1 - self.mix_ratio)
+        output_states = output_states + input_states
+
+        if not return_dict:
+            return (output_states,)
+
+        return Transformer2DModelOutput(sample=output_states)
+
+    def _set_attention_slice(self, slice_size):
+        for transformer in self.transformers:
+            transformer._set_attention_slice(slice_size)
+
+    def _set_use_memory_efficient_attention_xformers(self, use_memory_efficient_attention_xformers: bool):
+        for transformer in self.transformers:
+            transformer._set_use_memory_efficient_attention_xformers(use_memory_efficient_attention_xformers)

src/diffusers/models/unet_2d.py

@@ -175,7 +175,7 @@ class UNet2DModel(ModelMixin, ConfigMixin):
         num_groups_out = norm_num_groups if norm_num_groups is not None else min(block_out_channels[0] // 4, 32)
         self.conv_norm_out = nn.GroupNorm(num_channels=block_out_channels[0], num_groups=num_groups_out, eps=norm_eps)
         self.conv_act = nn.SiLU()
-        self.conv_out = nn.Conv2d(block_out_channels[0], out_channels, 3, padding=1)
+        self.conv_out = nn.Conv2d(block_out_channels[0], out_channels, kernel_size=3, padding=1)
 
     def forward(
         self,

src/diffusers/models/unet_2d_blocks.py

@@ -15,7 +15,7 @@ import numpy as np
 import torch
 from torch import nn
 
-from .attention import AttentionBlock, Transformer2DModel
+from .attention import AttentionBlock, DualTransformer2DModel, Transformer2DModel
 from .resnet import Downsample2D, FirDownsample2D, FirUpsample2D, ResnetBlock2D, Upsample2D
 
 
@@ -32,6 +32,7 @@ def get_down_block(
     resnet_groups=None,
     cross_attention_dim=None,
     downsample_padding=None,
+    dual_cross_attention=False,
 ):
     down_block_type = down_block_type[7:] if down_block_type.startswith("UNetRes") else down_block_type
     if down_block_type == "DownBlock2D":
@@ -74,6 +75,7 @@ def get_down_block(
             downsample_padding=downsample_padding,
             cross_attention_dim=cross_attention_dim,
             attn_num_head_channels=attn_num_head_channels,
+            dual_cross_attention=dual_cross_attention,
         )
     elif down_block_type == "SkipDownBlock2D":
         return SkipDownBlock2D(
@@ -137,6 +139,7 @@ def get_up_block(
     attn_num_head_channels,
     resnet_groups=None,
     cross_attention_dim=None,
+    dual_cross_attention=False,
 ):
     up_block_type = up_block_type[7:] if up_block_type.startswith("UNetRes") else up_block_type
     if up_block_type == "UpBlock2D":
@@ -166,6 +169,7 @@ def get_up_block(
             resnet_groups=resnet_groups,
             cross_attention_dim=cross_attention_dim,
             attn_num_head_channels=attn_num_head_channels,
+            dual_cross_attention=dual_cross_attention,
         )
     elif up_block_type == "AttnUpBlock2D":
         return AttnUpBlock2D(
@@ -322,6 +326,7 @@ class UNetMidBlock2DCrossAttn(nn.Module):
         attention_type="default",
         output_scale_factor=1.0,
         cross_attention_dim=1280,
+        dual_cross_attention=False,
         **kwargs,
     ):
         super().__init__()
@@ -348,16 +353,28 @@ class UNetMidBlock2DCrossAttn(nn.Module):
         attentions = []
 
         for _ in range(num_layers):
-            attentions.append(
-                Transformer2DModel(
-                    attn_num_head_channels,
-                    in_channels // attn_num_head_channels,
-                    in_channels=in_channels,
-                    num_layers=1,
-                    cross_attention_dim=cross_attention_dim,
-                    norm_num_groups=resnet_groups,
+            if not dual_cross_attention:
+                attentions.append(
+                    Transformer2DModel(
+                        attn_num_head_channels,
+                        in_channels // attn_num_head_channels,
+                        in_channels=in_channels,
+                        num_layers=1,
+                        cross_attention_dim=cross_attention_dim,
+                        norm_num_groups=resnet_groups,
+                    )
+                )
+            else:
+                attentions.append(
+                    DualTransformer2DModel(
+                        attn_num_head_channels,
+                        in_channels // attn_num_head_channels,
+                        in_channels=in_channels,
+                        num_layers=1,
+                        cross_attention_dim=cross_attention_dim,
+                        norm_num_groups=resnet_groups,
+                    )
                 )
-            )
             resnets.append(
                 ResnetBlock2D(
                     in_channels=in_channels,
@@ -505,6 +522,7 @@ class CrossAttnDownBlock2D(nn.Module):
         output_scale_factor=1.0,
         downsample_padding=1,
         add_downsample=True,
+        dual_cross_attention=False,
     ):
         super().__init__()
         resnets = []
@@ -529,16 +547,28 @@ class CrossAttnDownBlock2D(nn.Module):
                     pre_norm=resnet_pre_norm,
                 )
             )
-            attentions.append(
-                Transformer2DModel(
-                    attn_num_head_channels,
-                    out_channels // attn_num_head_channels,
-                    in_channels=out_channels,
-                    num_layers=1,
-                    cross_attention_dim=cross_attention_dim,
-                    norm_num_groups=resnet_groups,
+            if not dual_cross_attention:
+                attentions.append(
+                    Transformer2DModel(
+                        attn_num_head_channels,
+                        out_channels // attn_num_head_channels,
+                        in_channels=out_channels,
+                        num_layers=1,
+                        cross_attention_dim=cross_attention_dim,
+                        norm_num_groups=resnet_groups,
+                    )
+                )
+            else:
+                attentions.append(
+                    DualTransformer2DModel(
+                        attn_num_head_channels,
+                        out_channels // attn_num_head_channels,
+                        in_channels=out_channels,
+                        num_layers=1,
+                        cross_attention_dim=cross_attention_dim,
+                        norm_num_groups=resnet_groups,
+                    )
                 )
-            )
         self.attentions = nn.ModuleList(attentions)
         self.resnets = nn.ModuleList(resnets)
 
@@ -1089,6 +1119,7 @@ class CrossAttnUpBlock2D(nn.Module):
         attention_type="default",
         output_scale_factor=1.0,
         add_upsample=True,
+        dual_cross_attention=False,
     ):
         super().__init__()
         resnets = []
@@ -1115,16 +1146,28 @@ class CrossAttnUpBlock2D(nn.Module):
                     pre_norm=resnet_pre_norm,
                 )
             )
-            attentions.append(
-                Transformer2DModel(
-                    attn_num_head_channels,
-                    out_channels // attn_num_head_channels,
-                    in_channels=out_channels,
-                    num_layers=1,
-                    cross_attention_dim=cross_attention_dim,
-                    norm_num_groups=resnet_groups,
+            if not dual_cross_attention:
+                attentions.append(
+                    Transformer2DModel(
+                        attn_num_head_channels,
+                        out_channels // attn_num_head_channels,
+                        in_channels=out_channels,
+                        num_layers=1,
+                        cross_attention_dim=cross_attention_dim,
+                        norm_num_groups=resnet_groups,
+                    )
+                )
+            else:
+                attentions.append(
+                    DualTransformer2DModel(
+                        attn_num_head_channels,
+                        out_channels // attn_num_head_channels,
+                        in_channels=out_channels,
+                        num_layers=1,
+                        cross_attention_dim=cross_attention_dim,
+                        norm_num_groups=resnet_groups,
+                    )
                 )
-            )
         self.attentions = nn.ModuleList(attentions)
         self.resnets = nn.ModuleList(resnets)
 

src/diffusers/models/unet_2d_condition.py

@@ -107,6 +107,7 @@ class UNet2DConditionModel(ModelMixin, ConfigMixin):
         norm_eps: float = 1e-5,
         cross_attention_dim: int = 1280,
         attention_head_dim: int = 8,
+        dual_cross_attention: bool = False,
     ):
         super().__init__()
 
@@ -146,6 +147,7 @@ class UNet2DConditionModel(ModelMixin, ConfigMixin):
                 cross_attention_dim=cross_attention_dim,
                 attn_num_head_channels=attention_head_dim,
                 downsample_padding=downsample_padding,
+                dual_cross_attention=dual_cross_attention,
             )
             self.down_blocks.append(down_block)
 
@@ -160,6 +162,7 @@ class UNet2DConditionModel(ModelMixin, ConfigMixin):
             cross_attention_dim=cross_attention_dim,
             attn_num_head_channels=attention_head_dim,
             resnet_groups=norm_num_groups,
+            dual_cross_attention=dual_cross_attention,
         )
 
         # count how many layers upsample the images
@@ -195,6 +198,7 @@ class UNet2DConditionModel(ModelMixin, ConfigMixin):
                 resnet_groups=norm_num_groups,
                 cross_attention_dim=cross_attention_dim,
                 attn_num_head_channels=attention_head_dim,
+                dual_cross_attention=dual_cross_attention,
             )
             self.up_blocks.append(up_block)
             prev_output_channel = output_channel
@@ -202,7 +206,7 @@ class UNet2DConditionModel(ModelMixin, ConfigMixin):
         # out
         self.conv_norm_out = nn.GroupNorm(num_channels=block_out_channels[0], num_groups=norm_num_groups, eps=norm_eps)
         self.conv_act = nn.SiLU()
-        self.conv_out = nn.Conv2d(block_out_channels[0], out_channels, 3, padding=1)
+        self.conv_out = nn.Conv2d(block_out_channels[0], out_channels, kernel_size=3, padding=1)
 
     def set_attention_slice(self, slice_size):
         if slice_size is not None and self.config.attention_head_dim % slice_size != 0:

src/diffusers/pipelines/__init__.py

@@ -26,6 +26,12 @@ if is_torch_available() and is_transformers_available():
         StableDiffusionPipeline,
     )
     from .stable_diffusion_safe import StableDiffusionPipelineSafe
+    from .versatile_diffusion import (
+        VersatileDiffusionDualGuidedPipeline,
+        VersatileDiffusionImageVariationPipeline,
+        VersatileDiffusionPipeline,
+        VersatileDiffusionTextToImagePipeline,
+    )
     from .vq_diffusion import VQDiffusionPipeline
 
 if is_transformers_available() and is_onnx_available():

src/diffusers/pipelines/versatile_diffusion/__init__.py

+from ...utils import is_torch_available, is_transformers_available
+
+
+if is_transformers_available() and is_torch_available():
+    from .modeling_text_unet import UNetFlatConditionModel
+    from .pipeline_versatile_diffusion import VersatileDiffusionPipeline
+    from .pipeline_versatile_diffusion_dual_guided import VersatileDiffusionDualGuidedPipeline
+    from .pipeline_versatile_diffusion_image_variation import VersatileDiffusionImageVariationPipeline
+    from .pipeline_versatile_diffusion_text_to_image import VersatileDiffusionTextToImagePipeline

src/diffusers/utils/dummy_torch_and_transformers_objects.py

@@ -154,6 +154,66 @@ class StableDiffusionPipelineSafe(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class VersatileDiffusionDualGuidedPipeline(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 VersatileDiffusionImageVariationPipeline(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 VersatileDiffusionPipeline(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 VersatileDiffusionTextToImagePipeline(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 VQDiffusionPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 

tests/pipelines/versatile_diffusion/test_versatile_diffusion_dual_guided.py

+# 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 tempfile
+import unittest
+
+import numpy as np
+import torch
+
+from diffusers import VersatileDiffusionDualGuidedPipeline
+from diffusers.utils.testing_utils import load_image, require_torch_gpu, slow, torch_device
+
+from ...test_pipelines_common import PipelineTesterMixin
+
+
+torch.backends.cuda.matmul.allow_tf32 = False
+
+
+class VersatileDiffusionDualGuidedPipelineFastTests(PipelineTesterMixin, unittest.TestCase):
+    pass
+
+
+@slow
+@require_torch_gpu
+class VersatileDiffusionDualGuidedPipelineIntegrationTests(unittest.TestCase):
+    def tearDown(self):
+        # clean up the VRAM after each test
+        super().tearDown()
+        gc.collect()
+        torch.cuda.empty_cache()
+
+    def test_remove_unused_weights_save_load(self):
+        pipe = VersatileDiffusionDualGuidedPipeline.from_pretrained("shi-labs/versatile-diffusion")
+        # remove text_unet
+        pipe.remove_unused_weights()
+        pipe.to(torch_device)
+        pipe.set_progress_bar_config(disable=None)
+
+        second_prompt = load_image(
+            "https://raw.githubusercontent.com/SHI-Labs/Versatile-Diffusion/master/assets/benz.jpg"
+        )
+
+        generator = torch.Generator(device=torch_device).manual_seed(0)
+        image = pipe(
+            prompt="first prompt",
+            image=second_prompt,
+            text_to_image_strength=0.75,
+            generator=generator,
+            guidance_scale=7.5,
+            num_inference_steps=2,
+            output_type="numpy",
+        ).images
+
+        with tempfile.TemporaryDirectory() as tmpdirname:
+            pipe.save_pretrained(tmpdirname)
+            pipe = VersatileDiffusionDualGuidedPipeline.from_pretrained(tmpdirname)
+
+        pipe.to(torch_device)
+        pipe.set_progress_bar_config(disable=None)
+
+        generator = generator.manual_seed(0)
+        new_image = pipe(
+            prompt="first prompt",
+            image=second_prompt,
+            text_to_image_strength=0.75,
+            generator=generator,
+            guidance_scale=7.5,
+            num_inference_steps=2,
+            output_type="numpy",
+        ).images
+
+        assert np.abs(image - new_image).sum() < 1e-5, "Models don't have the same forward pass"
+
+    def test_inference_dual_guided(self):
+        pipe = VersatileDiffusionDualGuidedPipeline.from_pretrained("shi-labs/versatile-diffusion")
+        pipe.remove_unused_weights()
+        pipe.to(torch_device)
+        pipe.set_progress_bar_config(disable=None)
+
+        first_prompt = "cyberpunk 2077"
+        second_prompt = load_image(
+            "https://raw.githubusercontent.com/SHI-Labs/Versatile-Diffusion/master/assets/benz.jpg"
+        )
+        generator = torch.Generator(device=torch_device).manual_seed(0)
+        image = pipe(
+            prompt=first_prompt,
+            image=second_prompt,
+            text_to_image_strength=0.75,
+            generator=generator,
+            guidance_scale=7.5,
+            num_inference_steps=50,
+            output_type="numpy",
+        ).images
+
+        image_slice = image[0, 253:256, 253:256, -1]
+
+        assert image.shape == (1, 512, 512, 3)
+        expected_slice = np.array([0.014, 0.0112, 0.0136, 0.0145, 0.0107, 0.0113, 0.0272, 0.0215, 0.0216])
+        assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-2