diffusers 0.28.2__py3-none-any.whl → 0.29.1__py3-none-any.whl

diffusers/models/unets/unet_1d.py

@@ -206,11 +206,11 @@ class UNet1DModel(ModelMixin, ConfigMixin):
                 The noisy input tensor with the following shape `(batch_size, num_channels, sample_size)`.
             timestep (`torch.Tensor` or `float` or `int`): The number of timesteps to denoise an input.
             return_dict (`bool`, *optional*, defaults to `True`):
-                Whether or not to return a [`~models.unet_1d.UNet1DOutput`] instead of a plain tuple.
+                Whether or not to return a [`~models.unets.unet_1d.UNet1DOutput`] instead of a plain tuple.
 
         Returns:
-            [`~models.unet_1d.UNet1DOutput`] or `tuple`:
-                If `return_dict` is True, an [`~models.unet_1d.UNet1DOutput`] is returned, otherwise a `tuple` is
+            [`~models.unets.unet_1d.UNet1DOutput`] or `tuple`:
+                If `return_dict` is True, an [`~models.unets.unet_1d.UNet1DOutput`] is returned, otherwise a `tuple` is
                 returned where the first element is the sample tensor.
         """
 

diffusers/models/unets/unet_2d.py

@@ -257,11 +257,11 @@ class UNet2DModel(ModelMixin, ConfigMixin):
             class_labels (`torch.Tensor`, *optional*, defaults to `None`):
                 Optional class labels for conditioning. Their embeddings will be summed with the timestep embeddings.
             return_dict (`bool`, *optional*, defaults to `True`):
-                Whether or not to return a [`~models.unet_2d.UNet2DOutput`] instead of a plain tuple.
+                Whether or not to return a [`~models.unets.unet_2d.UNet2DOutput`] instead of a plain tuple.
 
         Returns:
-            [`~models.unet_2d.UNet2DOutput`] or `tuple`:
-                If `return_dict` is True, an [`~models.unet_2d.UNet2DOutput`] is returned, otherwise a `tuple` is
+            [`~models.unets.unet_2d.UNet2DOutput`] or `tuple`:
+                If `return_dict` is True, an [`~models.unets.unet_2d.UNet2DOutput`] is returned, otherwise a `tuple` is
                 returned where the first element is the sample tensor.
         """
         # 0. center input if necessary

diffusers/models/unets/unet_2d_condition.py

@@ -110,13 +110,13 @@ class UNet2DConditionModel(
             The dimension of the cross attention features.
         transformer_layers_per_block (`int`, `Tuple[int]`, or `Tuple[Tuple]` , *optional*, defaults to 1):
             The number of transformer blocks of type [`~models.attention.BasicTransformerBlock`]. Only relevant for
-            [`~models.unet_2d_blocks.CrossAttnDownBlock2D`], [`~models.unet_2d_blocks.CrossAttnUpBlock2D`],
-            [`~models.unet_2d_blocks.UNetMidBlock2DCrossAttn`].
+            [`~models.unets.unet_2d_blocks.CrossAttnDownBlock2D`], [`~models.unets.unet_2d_blocks.CrossAttnUpBlock2D`],
+            [`~models.unets.unet_2d_blocks.UNetMidBlock2DCrossAttn`].
         reverse_transformer_layers_per_block : (`Tuple[Tuple]`, *optional*, defaults to None):
             The number of transformer blocks of type [`~models.attention.BasicTransformerBlock`], in the upsampling
             blocks of the U-Net. Only relevant if `transformer_layers_per_block` is of type `Tuple[Tuple]` and for
-            [`~models.unet_2d_blocks.CrossAttnDownBlock2D`], [`~models.unet_2d_blocks.CrossAttnUpBlock2D`],
-            [`~models.unet_2d_blocks.UNetMidBlock2DCrossAttn`].
+            [`~models.unets.unet_2d_blocks.CrossAttnDownBlock2D`], [`~models.unets.unet_2d_blocks.CrossAttnUpBlock2D`],
+            [`~models.unets.unet_2d_blocks.UNetMidBlock2DCrossAttn`].
         encoder_hid_dim (`int`, *optional*, defaults to None):
             If `encoder_hid_dim_type` is defined, `encoder_hidden_states` will be projected from `encoder_hid_dim`
             dimension to `cross_attention_dim`.
@@ -903,17 +903,6 @@ class UNet2DConditionModel(
         if self.original_attn_processors is not None:
             self.set_attn_processor(self.original_attn_processors)
 
-    def unload_lora(self):
-        """Unloads LoRA weights."""
-        deprecate(
-            "unload_lora",
-            "0.28.0",
-            "Calling `unload_lora()` is deprecated and will be removed in a future version. Please install `peft` and then call `disable_adapters().",
-        )
-        for module in self.modules():
-            if hasattr(module, "set_lora_layer"):
-                module.set_lora_layer(None)
-
     def get_time_embed(
         self, sample: torch.Tensor, timestep: Union[torch.Tensor, float, int]
     ) -> Optional[torch.Tensor]:

diffusers/models/unets/unet_3d_condition.py

@@ -22,7 +22,7 @@ import torch.utils.checkpoint
 
 from ...configuration_utils import ConfigMixin, register_to_config
 from ...loaders import UNet2DConditionLoadersMixin
-from ...utils import BaseOutput, deprecate, logging
+from ...utils import BaseOutput, logging
 from ..activations import get_activation
 from ..attention_processor import (
     ADDED_KV_ATTENTION_PROCESSORS,
@@ -546,18 +546,6 @@ class UNet3DConditionModel(ModelMixin, ConfigMixin, UNet2DConditionLoadersMixin)
         if self.original_attn_processors is not None:
             self.set_attn_processor(self.original_attn_processors)
 
-    # Copied from diffusers.models.unets.unet_2d_condition.UNet2DConditionModel.unload_lora
-    def unload_lora(self):
-        """Unloads LoRA weights."""
-        deprecate(
-            "unload_lora",
-            "0.28.0",
-            "Calling `unload_lora()` is deprecated and will be removed in a future version. Please install `peft` and then call `disable_adapters().",
-        )
-        for module in self.modules():
-            if hasattr(module, "set_lora_layer"):
-                module.set_lora_layer(None)
-
     def forward(
         self,
         sample: torch.Tensor,
@@ -598,15 +586,15 @@ class UNet3DConditionModel(ModelMixin, ConfigMixin, UNet2DConditionLoadersMixin)
             mid_block_additional_residual: (`torch.Tensor`, *optional*):
                 A tensor that if specified is added to the residual of the middle unet block.
             return_dict (`bool`, *optional*, defaults to `True`):
-                Whether or not to return a [`~models.unet_3d_condition.UNet3DConditionOutput`] instead of a plain
+                Whether or not to return a [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] instead of a plain
                 tuple.
             cross_attention_kwargs (`dict`, *optional*):
                 A kwargs dictionary that if specified is passed along to the [`AttnProcessor`].
 
         Returns:
-            [`~models.unet_3d_condition.UNet3DConditionOutput`] or `tuple`:
-                If `return_dict` is True, an [`~models.unet_3d_condition.UNet3DConditionOutput`] is returned, otherwise
-                a `tuple` is returned where the first element is the sample tensor.
+            [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] or `tuple`:
+                If `return_dict` is True, an [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] is returned,
+                otherwise a `tuple` is returned where the first element is the sample tensor.
         """
         # By default samples have to be AT least a multiple of the overall upsampling factor.
         # The overall upsampling factor is equal to 2 ** (# num of upsampling layears).

diffusers/models/unets/unet_i2vgen_xl.py

@@ -542,13 +542,13 @@ class I2VGenXLUNet(ModelMixin, ConfigMixin, UNet2DConditionLoadersMixin):
                 `self.processor` in
                 [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
             return_dict (`bool`, *optional*, defaults to `True`):
-                Whether or not to return a [`~models.unet_3d_condition.UNet3DConditionOutput`] instead of a plain
+                Whether or not to return a [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] instead of a plain
                 tuple.
 
         Returns:
-            [`~models.unet_3d_condition.UNet3DConditionOutput`] or `tuple`:
-                If `return_dict` is True, an [`~models.unet_3d_condition.UNet3DConditionOutput`] is returned, otherwise
-                a `tuple` is returned where the first element is the sample tensor.
+            [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] or `tuple`:
+                If `return_dict` is True, an [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] is returned,
+                otherwise a `tuple` is returned where the first element is the sample tensor.
         """
         batch_size, channels, num_frames, height, width = sample.shape
 

diffusers/models/unets/unet_motion_model.py

@@ -856,13 +856,13 @@ class UNetMotionModel(ModelMixin, ConfigMixin, UNet2DConditionLoadersMixin):
             mid_block_additional_residual: (`torch.Tensor`, *optional*):
                 A tensor that if specified is added to the residual of the middle unet block.
             return_dict (`bool`, *optional*, defaults to `True`):
-                Whether or not to return a [`~models.unet_3d_condition.UNet3DConditionOutput`] instead of a plain
+                Whether or not to return a [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] instead of a plain
                 tuple.
 
         Returns:
-            [`~models.unet_3d_condition.UNet3DConditionOutput`] or `tuple`:
-                If `return_dict` is True, an [`~models.unet_3d_condition.UNet3DConditionOutput`] is returned, otherwise
-                a `tuple` is returned where the first element is the sample tensor.
+            [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] or `tuple`:
+                If `return_dict` is True, an [`~models.unets.unet_3d_condition.UNet3DConditionOutput`] is returned,
+                otherwise a `tuple` is returned where the first element is the sample tensor.
         """
         # By default samples have to be AT least a multiple of the overall upsampling factor.
         # The overall upsampling factor is equal to 2 ** (# num of upsampling layears).

diffusers/models/unets/unet_spatio_temporal_condition.py

@@ -57,9 +57,9 @@ class UNetSpatioTemporalConditionModel(ModelMixin, ConfigMixin, UNet2DConditionL
             The dimension of the cross attention features.
         transformer_layers_per_block (`int`, `Tuple[int]`, or `Tuple[Tuple]` , *optional*, defaults to 1):
             The number of transformer blocks of type [`~models.attention.BasicTransformerBlock`]. Only relevant for
-            [`~models.unet_3d_blocks.CrossAttnDownBlockSpatioTemporal`],
-            [`~models.unet_3d_blocks.CrossAttnUpBlockSpatioTemporal`],
-            [`~models.unet_3d_blocks.UNetMidBlockSpatioTemporal`].
+            [`~models.unets.unet_3d_blocks.CrossAttnDownBlockSpatioTemporal`],
+            [`~models.unets.unet_3d_blocks.CrossAttnUpBlockSpatioTemporal`],
+            [`~models.unets.unet_3d_blocks.UNetMidBlockSpatioTemporal`].
         num_attention_heads (`int`, `Tuple[int]`, defaults to `(5, 10, 10, 20)`):
             The number of attention heads.
         dropout (`float`, *optional*, defaults to 0.0): The dropout probability to use.

diffusers/models/vq_model.py

@@ -11,172 +11,15 @@
 # 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.
-from dataclasses import dataclass
-from typing import Optional, Tuple, Union
+from ..utils import deprecate
+from .autoencoders.vq_model import VQEncoderOutput, VQModel
 
-import torch
-import torch.nn as nn
 
-from ..configuration_utils import ConfigMixin, register_to_config
-from ..utils import BaseOutput
-from ..utils.accelerate_utils import apply_forward_hook
-from .autoencoders.vae import Decoder, DecoderOutput, Encoder, VectorQuantizer
-from .modeling_utils import ModelMixin
+class VQEncoderOutput(VQEncoderOutput):
+    deprecation_message = "Importing `VQEncoderOutput` from `diffusers.models.vq_model` is deprecated and this will be removed in a future version. Please use `from diffusers.models.autoencoders.vq_model import VQEncoderOutput`, instead."
+    deprecate("VQEncoderOutput", "0.31", deprecation_message)
 
 
-@dataclass
-class VQEncoderOutput(BaseOutput):
-    """
-    Output of VQModel encoding method.
-
-    Args:
-        latents (`torch.Tensor` of shape `(batch_size, num_channels, height, width)`):
-            The encoded output sample from the last layer of the model.
-    """
-
-    latents: torch.Tensor
-
-
-class VQModel(ModelMixin, ConfigMixin):
-    r"""
-    A VQ-VAE model for decoding latent representations.
-
-    This model inherits from [`ModelMixin`]. Check the superclass documentation for it's generic methods implemented
-    for all models (such as downloading or saving).
-
-    Parameters:
-        in_channels (int, *optional*, defaults to 3): Number of channels in the input image.
-        out_channels (int,  *optional*, defaults to 3): Number of channels in the output.
-        down_block_types (`Tuple[str]`, *optional*, defaults to `("DownEncoderBlock2D",)`):
-            Tuple of downsample block types.
-        up_block_types (`Tuple[str]`, *optional*, defaults to `("UpDecoderBlock2D",)`):
-            Tuple of upsample block types.
-        block_out_channels (`Tuple[int]`, *optional*, defaults to `(64,)`):
-            Tuple of block output channels.
-        layers_per_block (`int`, *optional*, defaults to `1`): Number of layers per block.
-        act_fn (`str`, *optional*, defaults to `"silu"`): The activation function to use.
-        latent_channels (`int`, *optional*, defaults to `3`): Number of channels in the latent space.
-        sample_size (`int`, *optional*, defaults to `32`): Sample input size.
-        num_vq_embeddings (`int`, *optional*, defaults to `256`): Number of codebook vectors in the VQ-VAE.
-        norm_num_groups (`int`, *optional*, defaults to `32`): Number of groups for normalization layers.
-        vq_embed_dim (`int`, *optional*): Hidden dim of codebook vectors in the VQ-VAE.
-        scaling_factor (`float`, *optional*, defaults to `0.18215`):
-            The component-wise standard deviation of the trained latent space computed using the first batch of the
-            training set. This is used to scale the latent space to have unit variance when training the diffusion
-            model. The latents are scaled with the formula `z = z * scaling_factor` before being passed to the
-            diffusion model. When decoding, the latents are scaled back to the original scale with the formula: `z = 1
-            / scaling_factor * z`. For more details, refer to sections 4.3.2 and D.1 of the [High-Resolution Image
-            Synthesis with Latent Diffusion Models](https://arxiv.org/abs/2112.10752) paper.
-        norm_type (`str`, *optional*, defaults to `"group"`):
-            Type of normalization layer to use. Can be one of `"group"` or `"spatial"`.
-    """
-
-    @register_to_config
-    def __init__(
-        self,
-        in_channels: int = 3,
-        out_channels: int = 3,
-        down_block_types: Tuple[str, ...] = ("DownEncoderBlock2D",),
-        up_block_types: Tuple[str, ...] = ("UpDecoderBlock2D",),
-        block_out_channels: Tuple[int, ...] = (64,),
-        layers_per_block: int = 1,
-        act_fn: str = "silu",
-        latent_channels: int = 3,
-        sample_size: int = 32,
-        num_vq_embeddings: int = 256,
-        norm_num_groups: int = 32,
-        vq_embed_dim: Optional[int] = None,
-        scaling_factor: float = 0.18215,
-        norm_type: str = "group",  # group, spatial
-        mid_block_add_attention=True,
-        lookup_from_codebook=False,
-        force_upcast=False,
-    ):
-        super().__init__()
-
-        # pass init params to Encoder
-        self.encoder = Encoder(
-            in_channels=in_channels,
-            out_channels=latent_channels,
-            down_block_types=down_block_types,
-            block_out_channels=block_out_channels,
-            layers_per_block=layers_per_block,
-            act_fn=act_fn,
-            norm_num_groups=norm_num_groups,
-            double_z=False,
-            mid_block_add_attention=mid_block_add_attention,
-        )
-
-        vq_embed_dim = vq_embed_dim if vq_embed_dim is not None else latent_channels
-
-        self.quant_conv = nn.Conv2d(latent_channels, vq_embed_dim, 1)
-        self.quantize = VectorQuantizer(num_vq_embeddings, vq_embed_dim, beta=0.25, remap=None, sane_index_shape=False)
-        self.post_quant_conv = nn.Conv2d(vq_embed_dim, latent_channels, 1)
-
-        # pass init params to Decoder
-        self.decoder = Decoder(
-            in_channels=latent_channels,
-            out_channels=out_channels,
-            up_block_types=up_block_types,
-            block_out_channels=block_out_channels,
-            layers_per_block=layers_per_block,
-            act_fn=act_fn,
-            norm_num_groups=norm_num_groups,
-            norm_type=norm_type,
-            mid_block_add_attention=mid_block_add_attention,
-        )
-
-    @apply_forward_hook
-    def encode(self, x: torch.Tensor, return_dict: bool = True) -> VQEncoderOutput:
-        h = self.encoder(x)
-        h = self.quant_conv(h)
-
-        if not return_dict:
-            return (h,)
-
-        return VQEncoderOutput(latents=h)
-
-    @apply_forward_hook
-    def decode(
-        self, h: torch.Tensor, force_not_quantize: bool = False, return_dict: bool = True, shape=None
-    ) -> Union[DecoderOutput, torch.Tensor]:
-        # also go through quantization layer
-        if not force_not_quantize:
-            quant, commit_loss, _ = self.quantize(h)
-        elif self.config.lookup_from_codebook:
-            quant = self.quantize.get_codebook_entry(h, shape)
-            commit_loss = torch.zeros((h.shape[0])).to(h.device, dtype=h.dtype)
-        else:
-            quant = h
-            commit_loss = torch.zeros((h.shape[0])).to(h.device, dtype=h.dtype)
-        quant2 = self.post_quant_conv(quant)
-        dec = self.decoder(quant2, quant if self.config.norm_type == "spatial" else None)
-
-        if not return_dict:
-            return dec, commit_loss
-
-        return DecoderOutput(sample=dec, commit_loss=commit_loss)
-
-    def forward(
-        self, sample: torch.Tensor, return_dict: bool = True
-    ) -> Union[DecoderOutput, Tuple[torch.Tensor, ...]]:
-        r"""
-        The [`VQModel`] forward method.
-
-        Args:
-            sample (`torch.Tensor`): Input sample.
-            return_dict (`bool`, *optional*, defaults to `True`):
-                Whether or not to return a [`models.vq_model.VQEncoderOutput`] instead of a plain tuple.
-
-        Returns:
-            [`~models.vq_model.VQEncoderOutput`] or `tuple`:
-                If return_dict is True, a [`~models.vq_model.VQEncoderOutput`] is returned, otherwise a plain `tuple`
-                is returned.
-        """
-
-        h = self.encode(sample).latents
-        dec = self.decode(h)
-
-        if not return_dict:
-            return dec.sample, dec.commit_loss
-        return dec
+class VQModel(VQModel):
+    deprecation_message = "Importing `VQModel` from `diffusers.models.vq_model` is deprecated and this will be removed in a future version. Please use `from diffusers.models.autoencoders.vq_model import VQModel`, instead."
+    deprecate("VQModel", "0.31", deprecation_message)

diffusers/pipelines/__init__.py

@@ -20,6 +20,7 @@ from ..utils import (
 _dummy_objects = {}
 _import_structure = {
     "controlnet": [],
+    "controlnet_sd3": [],
     "controlnet_xs": [],
     "deprecated": [],
     "latent_diffusion": [],
@@ -142,6 +143,11 @@ else:
             "StableDiffusionXLControlNetXSPipeline",
         ]
     )
+    _import_structure["controlnet_sd3"].extend(
+        [
+            "StableDiffusion3ControlNetPipeline",
+        ]
+    )
     _import_structure["deepfloyd_if"] = [
         "IFImg2ImgPipeline",
         "IFImg2ImgSuperResolutionPipeline",
@@ -220,6 +226,7 @@ else:
             "StableDiffusionLDM3DPipeline",
         ]
     )
+    _import_structure["stable_diffusion_3"] = ["StableDiffusion3Pipeline", "StableDiffusion3Img2ImgPipeline"]
     _import_structure["stable_diffusion_attend_and_excite"] = ["StableDiffusionAttendAndExcitePipeline"]
     _import_structure["stable_diffusion_safe"] = ["StableDiffusionPipelineSafe"]
     _import_structure["stable_diffusion_sag"] = ["StableDiffusionSAGPipeline"]
@@ -393,6 +400,9 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             StableDiffusionXLControlNetInpaintPipeline,
             StableDiffusionXLControlNetPipeline,
         )
+        from .controlnet_sd3 import (
+            StableDiffusion3ControlNetPipeline,
+        )
         from .controlnet_xs import (
             StableDiffusionControlNetXSPipeline,
             StableDiffusionXLControlNetXSPipeline,
@@ -485,6 +495,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             StableUnCLIPImg2ImgPipeline,
             StableUnCLIPPipeline,
         )
+        from .stable_diffusion_3 import StableDiffusion3Img2ImgPipeline, StableDiffusion3Pipeline
         from .stable_diffusion_attend_and_excite import StableDiffusionAttendAndExcitePipeline
         from .stable_diffusion_diffedit import StableDiffusionDiffEditPipeline
         from .stable_diffusion_gligen import StableDiffusionGLIGENPipeline, StableDiffusionGLIGENTextImagePipeline

diffusers/pipelines/animatediff/pipeline_animatediff.py

@@ -316,9 +316,10 @@ class AnimateDiffPipeline(
             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)
 
-        if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
-            # Retrieve the original scale by scaling back the LoRA layers
-            unscale_lora_layers(self.text_encoder, lora_scale)
+        if self.text_encoder is not None:
+            if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
+                # Retrieve the original scale by scaling back the LoRA layers
+                unscale_lora_layers(self.text_encoder, lora_scale)
 
         return prompt_embeds, negative_prompt_embeds
 

diffusers/pipelines/animatediff/pipeline_animatediff_video2video.py

@@ -420,9 +420,10 @@ class AnimateDiffVideoToVideoPipeline(
             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)
 
-        if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
-            # Retrieve the original scale by scaling back the LoRA layers
-            unscale_lora_layers(self.text_encoder, lora_scale)
+        if self.text_encoder is not None:
+            if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
+                # Retrieve the original scale by scaling back the LoRA layers
+                unscale_lora_layers(self.text_encoder, lora_scale)
 
         return prompt_embeds, negative_prompt_embeds
 

diffusers/pipelines/auto_pipeline.py

@@ -27,6 +27,7 @@ from .controlnet import (
     StableDiffusionXLControlNetPipeline,
 )
 from .deepfloyd_if import IFImg2ImgPipeline, IFInpaintingPipeline, IFPipeline
+from .hunyuandit import HunyuanDiTPipeline
 from .kandinsky import (
     KandinskyCombinedPipeline,
     KandinskyImg2ImgCombinedPipeline,
@@ -52,6 +53,10 @@ from .stable_diffusion import (
     StableDiffusionInpaintPipeline,
     StableDiffusionPipeline,
 )
+from .stable_diffusion_3 import (
+    StableDiffusion3Img2ImgPipeline,
+    StableDiffusion3Pipeline,
+)
 from .stable_diffusion_xl import (
     StableDiffusionXLImg2ImgPipeline,
     StableDiffusionXLInpaintPipeline,
@@ -64,7 +69,9 @@ AUTO_TEXT2IMAGE_PIPELINES_MAPPING = OrderedDict(
     [
         ("stable-diffusion", StableDiffusionPipeline),
         ("stable-diffusion-xl", StableDiffusionXLPipeline),
+        ("stable-diffusion-3", StableDiffusion3Pipeline),
         ("if", IFPipeline),
+        ("hunyuan", HunyuanDiTPipeline),
         ("kandinsky", KandinskyCombinedPipeline),
         ("kandinsky22", KandinskyV22CombinedPipeline),
         ("kandinsky3", Kandinsky3Pipeline),
@@ -82,6 +89,7 @@ AUTO_IMAGE2IMAGE_PIPELINES_MAPPING = OrderedDict(
     [
         ("stable-diffusion", StableDiffusionImg2ImgPipeline),
         ("stable-diffusion-xl", StableDiffusionXLImg2ImgPipeline),
+        ("stable-diffusion-3", StableDiffusion3Img2ImgPipeline),
         ("if", IFImg2ImgPipeline),
         ("kandinsky", KandinskyImg2ImgCombinedPipeline),
         ("kandinsky22", KandinskyV22Img2ImgCombinedPipeline),

diffusers/pipelines/controlnet/pipeline_controlnet.py

@@ -463,9 +463,10 @@ class StableDiffusionControlNetPipeline(
             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)
 
-        if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
-            # Retrieve the original scale by scaling back the LoRA layers
-            unscale_lora_layers(self.text_encoder, lora_scale)
+        if self.text_encoder is not None:
+            if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
+                # Retrieve the original scale by scaling back the LoRA layers
+                unscale_lora_layers(self.text_encoder, lora_scale)
 
         return prompt_embeds, negative_prompt_embeds
 

diffusers/pipelines/controlnet/pipeline_controlnet_img2img.py

@@ -441,9 +441,10 @@ class StableDiffusionControlNetImg2ImgPipeline(
             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)
 
-        if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
-            # Retrieve the original scale by scaling back the LoRA layers
-            unscale_lora_layers(self.text_encoder, lora_scale)
+        if self.text_encoder is not None:
+            if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
+                # Retrieve the original scale by scaling back the LoRA layers
+                unscale_lora_layers(self.text_encoder, lora_scale)
 
         return prompt_embeds, negative_prompt_embeds
 

diffusers/pipelines/controlnet/pipeline_controlnet_inpaint.py

@@ -566,9 +566,10 @@ class StableDiffusionControlNetInpaintPipeline(
             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)
 
-        if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
-            # Retrieve the original scale by scaling back the LoRA layers
-            unscale_lora_layers(self.text_encoder, lora_scale)
+        if self.text_encoder is not None:
+            if isinstance(self, LoraLoaderMixin) and USE_PEFT_BACKEND:
+                # Retrieve the original scale by scaling back the LoRA layers
+                unscale_lora_layers(self.text_encoder, lora_scale)
 
         return prompt_embeds, negative_prompt_embeds
 

diffusers/pipelines/controlnet_sd3/__init__.py

@@ -0,0 +1,53 @@
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    DIFFUSERS_SLOW_IMPORT,
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    get_objects_from_module,
+    is_flax_available,
+    is_torch_available,
+    is_transformers_available,
+)
+
+
+_dummy_objects = {}
+_import_structure = {}
+
+try:
+    if not (is_transformers_available() and is_torch_available()):
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    from ...utils import dummy_torch_and_transformers_objects  # noqa F403
+
+    _dummy_objects.update(get_objects_from_module(dummy_torch_and_transformers_objects))
+else:
+    _import_structure["pipeline_stable_diffusion_3_controlnet"] = ["StableDiffusion3ControlNetPipeline"]
+
+if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
+    try:
+        if not (is_transformers_available() and is_torch_available()):
+            raise OptionalDependencyNotAvailable()
+
+    except OptionalDependencyNotAvailable:
+        from ...utils.dummy_torch_and_transformers_objects import *
+    else:
+        from .pipeline_stable_diffusion_3_controlnet import StableDiffusion3ControlNetPipeline
+
+    try:
+        if not (is_transformers_available() and is_flax_available()):
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        from ...utils.dummy_flax_and_transformers_objects import *  # noqa F403
+
+else:
+    import sys
+
+    sys.modules[__name__] = _LazyModule(
+        __name__,
+        globals()["__file__"],
+        _import_structure,
+        module_spec=__spec__,
+    )
+    for name, value in _dummy_objects.items():
+        setattr(sys.modules[__name__], name, value)