diffusers 0.27.0__py3-none-any.whl → 0.32.2__py3-none-any.whl

diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl_instruct_pix2pix.py

@@ -25,8 +25,6 @@ from ...models import AutoencoderKL, UNet2DConditionModel
 from ...models.attention_processor import (
     AttnProcessor2_0,
     FusedAttnProcessor2_0,
-    LoRAAttnProcessor2_0,
-    LoRAXFormersAttnProcessor,
     XFormersAttnProcessor,
 )
 from ...models.lora import adjust_lora_scale_text_encoder
@@ -169,6 +167,8 @@ class StableDiffusionXLInstructPix2PixPipeline(
             Whether to use the [invisible_watermark library](https://github.com/ShieldMnt/invisible-watermark/) to
             watermark output images. If not defined, it will default to True if the package is installed, otherwise no
             watermarker will be used.
+        is_cosxl_edit (`bool`, *optional*):
+            When set the image latents are scaled.
     """
 
     model_cpu_offload_seq = "text_encoder->text_encoder_2->unet->vae"
@@ -185,6 +185,7 @@ class StableDiffusionXLInstructPix2PixPipeline(
         scheduler: KarrasDiffusionSchedulers,
         force_zeros_for_empty_prompt: bool = True,
         add_watermarker: Optional[bool] = None,
+        is_cosxl_edit: Optional[bool] = False,
     ):
         super().__init__()
 
@@ -201,6 +202,7 @@ class StableDiffusionXLInstructPix2PixPipeline(
         self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
         self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
         self.default_sample_size = self.unet.config.sample_size
+        self.is_cosxl_edit = is_cosxl_edit
 
         add_watermarker = add_watermarker if add_watermarker is not None else is_invisible_watermark_available()
 
@@ -218,10 +220,10 @@ class StableDiffusionXLInstructPix2PixPipeline(
         do_classifier_free_guidance: bool = True,
         negative_prompt: Optional[str] = None,
         negative_prompt_2: Optional[str] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
-        pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        pooled_prompt_embeds: Optional[torch.Tensor] = None,
+        negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
         lora_scale: Optional[float] = None,
     ):
         r"""
@@ -246,17 +248,17 @@ class StableDiffusionXLInstructPix2PixPipeline(
             negative_prompt_2 (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
                 `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *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*):
+            negative_prompt_embeds (`torch.Tensor`, *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.
-            pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+            pooled_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
                 If not provided, pooled text embeddings will be generated from `prompt` input argument.
-            negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
                 input argument.
@@ -432,7 +434,6 @@ class StableDiffusionXLInstructPix2PixPipeline(
             extra_step_kwargs["generator"] = generator
         return extra_step_kwargs
 
-    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_instruct_pix2pix.StableDiffusionInstructPix2PixPipeline.check_inputs
     def check_inputs(
         self,
         prompt,
@@ -483,7 +484,12 @@ class StableDiffusionXLInstructPix2PixPipeline(
 
     # 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)
+        shape = (
+            batch_size,
+            num_channels_latents,
+            int(height) // self.vae_scale_factor,
+            int(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"
@@ -517,8 +523,8 @@ class StableDiffusionXLInstructPix2PixPipeline(
             # make sure the VAE is in float32 mode, as it overflows in float16
             needs_upcasting = self.vae.dtype == torch.float16 and self.vae.config.force_upcast
             if needs_upcasting:
+                image = image.float()
                 self.upcast_vae()
-                image = image.to(next(iter(self.vae.post_quant_conv.parameters())).dtype)
 
             image_latents = retrieve_latents(self.vae.encode(image), sample_mode="argmax")
 
@@ -551,6 +557,9 @@ class StableDiffusionXLInstructPix2PixPipeline(
         if image_latents.dtype != self.vae.dtype:
             image_latents = image_latents.to(dtype=self.vae.dtype)
 
+        if self.is_cosxl_edit:
+            image_latents = image_latents * self.vae.config.scaling_factor
+
         return image_latents
 
     # Copied from diffusers.pipelines.stable_diffusion_xl.pipeline_stable_diffusion_xl.StableDiffusionXLPipeline._get_add_time_ids
@@ -581,8 +590,6 @@ class StableDiffusionXLInstructPix2PixPipeline(
             (
                 AttnProcessor2_0,
                 XFormersAttnProcessor,
-                LoRAXFormersAttnProcessor,
-                LoRAAttnProcessor2_0,
                 FusedAttnProcessor2_0,
             ),
         )
@@ -611,14 +618,14 @@ class StableDiffusionXLInstructPix2PixPipeline(
         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,
-        pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
+        latents: Optional[torch.Tensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        pooled_prompt_embeds: Optional[torch.Tensor] = None,
+        negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
         output_type: Optional[str] = "pil",
         return_dict: bool = True,
-        callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
         callback_steps: int = 1,
         cross_attention_kwargs: Optional[Dict[str, Any]] = None,
         guidance_rescale: float = 0.0,
@@ -636,7 +643,7 @@ class StableDiffusionXLInstructPix2PixPipeline(
             prompt_2 (`str` or `List[str]`, *optional*):
                 The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
                 used in both text-encoders
-            image (`torch.FloatTensor` or `PIL.Image.Image` or `np.ndarray` or `List[torch.FloatTensor]` or `List[PIL.Image.Image]` or `List[np.ndarray]`):
+            image (`torch.Tensor` or `PIL.Image.Image` or `np.ndarray` or `List[torch.Tensor]` or `List[PIL.Image.Image]` or `List[np.ndarray]`):
                 The image(s) to modify with the pipeline.
             height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
                 The height in pixels of the generated image.
@@ -659,7 +666,7 @@ class StableDiffusionXLInstructPix2PixPipeline(
                 1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
                 usually at the expense of lower image quality.
             image_guidance_scale (`float`, *optional*, defaults to 1.5):
-                Image guidance scale is to push the generated image towards the inital image `image`. Image guidance
+                Image guidance scale is to push the generated image towards the initial image `image`. Image guidance
                 scale is enabled by setting `image_guidance_scale > 1`. Higher image guidance scale encourages to
                 generate images that are closely linked to the source image `image`, usually at the expense of lower
                 image quality. This pipeline requires a value of at least `1`.
@@ -678,21 +685,21 @@ class StableDiffusionXLInstructPix2PixPipeline(
             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*):
+            latents (`torch.Tensor`, *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*):
+            prompt_embeds (`torch.Tensor`, *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*):
+            negative_prompt_embeds (`torch.Tensor`, *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.
-            pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+            pooled_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
                 If not provided, pooled text embeddings will be generated from `prompt` input argument.
-            negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
                 input argument.
@@ -704,7 +711,7 @@ class StableDiffusionXLInstructPix2PixPipeline(
                 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)`.
+                called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
             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.
@@ -918,7 +925,12 @@ class StableDiffusionXLInstructPix2PixPipeline(
                     noise_pred = rescale_noise_cfg(noise_pred, noise_pred_text, guidance_rescale=guidance_rescale)
 
                 # compute the previous noisy sample x_t -> x_t-1
+                latents_dtype = latents.dtype
                 latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs, return_dict=False)[0]
+                if latents.dtype != latents_dtype:
+                    if torch.backends.mps.is_available():
+                        # some platforms (eg. apple mps) misbehave due to a pytorch bug: https://github.com/pytorch/pytorch/pull/99272
+                        latents = latents.to(latents_dtype)
 
                 # call the callback, if provided
                 if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
@@ -937,6 +949,10 @@ class StableDiffusionXLInstructPix2PixPipeline(
             if needs_upcasting:
                 self.upcast_vae()
                 latents = latents.to(next(iter(self.vae.post_quant_conv.parameters())).dtype)
+            elif latents.dtype != self.vae.dtype:
+                if torch.backends.mps.is_available():
+                    # some platforms (eg. apple mps) misbehave due to a pytorch bug: https://github.com/pytorch/pytorch/pull/99272
+                    self.vae = self.vae.to(latents.dtype)
 
             # unscale/denormalize the latents
             # denormalize with the mean and std if available and not None

diffusers/pipelines/stable_diffusion_xl/watermark.py

@@ -21,16 +21,22 @@ class StableDiffusionXLWatermarker:
 
         self.encoder.set_watermark("bits", self.watermark)
 
-    def apply_watermark(self, images: torch.FloatTensor):
+    def apply_watermark(self, images: torch.Tensor):
         # can't encode images that are smaller than 256
         if images.shape[-1] < 256:
             return images
 
         images = (255 * (images / 2 + 0.5)).cpu().permute(0, 2, 3, 1).float().numpy()
 
-        images = [self.encoder.encode(image, "dwtDct") for image in images]
+        # Convert RGB to BGR, which is the channel order expected by the watermark encoder.
+        images = images[:, :, :, ::-1]
 
-        images = torch.from_numpy(np.array(images)).permute(0, 3, 1, 2)
+        # Add watermark and convert BGR back to RGB
+        images = [self.encoder.encode(image, "dwtDct")[:, :, ::-1] for image in images]
+
+        images = np.array(images)
+
+        images = torch.from_numpy(images).permute(0, 3, 1, 2)
 
         images = torch.clamp(2 * (images / 255 - 0.5), min=-1.0, max=1.0)
         return images

diffusers/pipelines/stable_video_diffusion/pipeline_stable_video_diffusion.py

@@ -21,11 +21,12 @@ import PIL.Image
 import torch
 from transformers import CLIPImageProcessor, CLIPVisionModelWithProjection
 
-from ...image_processor import PipelineImageInput, VaeImageProcessor
+from ...image_processor import PipelineImageInput
 from ...models import AutoencoderKLTemporalDecoder, UNetSpatioTemporalConditionModel
 from ...schedulers import EulerDiscreteScheduler
 from ...utils import BaseOutput, logging, replace_example_docstring
 from ...utils.torch_utils import is_compiled_module, randn_tensor
+from ...video_processor import VideoProcessor
 from ..pipeline_utils import DiffusionPipeline
 
 
@@ -37,10 +38,14 @@ EXAMPLE_DOC_STRING = """
         >>> from diffusers import StableVideoDiffusionPipeline
         >>> from diffusers.utils import load_image, export_to_video
 
-        >>> pipe = StableVideoDiffusionPipeline.from_pretrained("stabilityai/stable-video-diffusion-img2vid-xt", torch_dtype=torch.float16, variant="fp16")
+        >>> pipe = StableVideoDiffusionPipeline.from_pretrained(
+        ...     "stabilityai/stable-video-diffusion-img2vid-xt", torch_dtype=torch.float16, variant="fp16"
+        ... )
         >>> pipe.to("cuda")
 
-        >>> image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/svd-docstring-example.jpeg")
+        >>> image = load_image(
+        ...     "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/svd-docstring-example.jpeg"
+        ... )
         >>> image = image.resize((1024, 576))
 
         >>> frames = pipe(image, num_frames=25, decode_chunk_size=8).frames[0]
@@ -57,26 +62,64 @@ def _append_dims(x, target_dims):
     return x[(...,) + (None,) * dims_to_append]
 
 
-# Copied from diffusers.pipelines.animatediff.pipeline_animatediff.tensor2vid
-def tensor2vid(video: torch.Tensor, processor: VaeImageProcessor, output_type: str = "np"):
-    batch_size, channels, num_frames, height, width = video.shape
-    outputs = []
-    for batch_idx in range(batch_size):
-        batch_vid = video[batch_idx].permute(1, 0, 2, 3)
-        batch_output = processor.postprocess(batch_vid, output_type)
-
-        outputs.append(batch_output)
-
-    if output_type == "np":
-        outputs = np.stack(outputs)
-
-    elif output_type == "pt":
-        outputs = torch.stack(outputs)
-
-    elif not output_type == "pil":
-        raise ValueError(f"{output_type} does not exist. Please choose one of ['np', 'pt', 'pil']")
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.retrieve_timesteps
+def retrieve_timesteps(
+    scheduler,
+    num_inference_steps: Optional[int] = None,
+    device: Optional[Union[str, torch.device]] = None,
+    timesteps: Optional[List[int]] = None,
+    sigmas: Optional[List[float]] = None,
+    **kwargs,
+):
+    r"""
+    Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles
+    custom timesteps. Any kwargs will be supplied to `scheduler.set_timesteps`.
 
-    return outputs
+    Args:
+        scheduler (`SchedulerMixin`):
+            The scheduler to get timesteps from.
+        num_inference_steps (`int`):
+            The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps`
+            must be `None`.
+        device (`str` or `torch.device`, *optional*):
+            The device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
+        timesteps (`List[int]`, *optional*):
+            Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed,
+            `num_inference_steps` and `sigmas` must be `None`.
+        sigmas (`List[float]`, *optional*):
+            Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed,
+            `num_inference_steps` and `timesteps` must be `None`.
+
+    Returns:
+        `Tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the
+        second element is the number of inference steps.
+    """
+    if timesteps is not None and sigmas is not None:
+        raise ValueError("Only one of `timesteps` or `sigmas` can be passed. Please choose one to set custom values")
+    if timesteps is not None:
+        accepts_timesteps = "timesteps" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+        if not accepts_timesteps:
+            raise ValueError(
+                f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+                f" timestep schedules. Please check whether you are using the correct scheduler."
+            )
+        scheduler.set_timesteps(timesteps=timesteps, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+        num_inference_steps = len(timesteps)
+    elif sigmas is not None:
+        accept_sigmas = "sigmas" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+        if not accept_sigmas:
+            raise ValueError(
+                f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+                f" sigmas schedules. Please check whether you are using the correct scheduler."
+            )
+        scheduler.set_timesteps(sigmas=sigmas, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+        num_inference_steps = len(timesteps)
+    else:
+        scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+    return timesteps, num_inference_steps
 
 
 @dataclass
@@ -85,12 +128,12 @@ class StableVideoDiffusionPipelineOutput(BaseOutput):
     Output class for Stable Video Diffusion pipeline.
 
     Args:
-        frames (`[List[List[PIL.Image.Image]]`, `np.ndarray`, `torch.FloatTensor`]):
-            List of denoised PIL images of length `batch_size` or numpy array or torch tensor
-            of shape `(batch_size, num_frames, height, width, num_channels)`.
+        frames (`[List[List[PIL.Image.Image]]`, `np.ndarray`, `torch.Tensor`]):
+            List of denoised PIL images of length `batch_size` or numpy array or torch tensor of shape `(batch_size,
+            num_frames, height, width, num_channels)`.
     """
 
-    frames: Union[List[List[PIL.Image.Image]], np.ndarray, torch.FloatTensor]
+    frames: Union[List[List[PIL.Image.Image]], np.ndarray, torch.Tensor]
 
 
 class StableVideoDiffusionPipeline(DiffusionPipeline):
@@ -104,7 +147,8 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
         vae ([`AutoencoderKLTemporalDecoder`]):
             Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations.
         image_encoder ([`~transformers.CLIPVisionModelWithProjection`]):
-            Frozen CLIP image-encoder ([laion/CLIP-ViT-H-14-laion2B-s32B-b79K](https://huggingface.co/laion/CLIP-ViT-H-14-laion2B-s32B-b79K)).
+            Frozen CLIP image-encoder
+            ([laion/CLIP-ViT-H-14-laion2B-s32B-b79K](https://huggingface.co/laion/CLIP-ViT-H-14-laion2B-s32B-b79K)).
         unet ([`UNetSpatioTemporalConditionModel`]):
             A `UNetSpatioTemporalConditionModel` to denoise the encoded image latents.
         scheduler ([`EulerDiscreteScheduler`]):
@@ -134,7 +178,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
             feature_extractor=feature_extractor,
         )
         self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
-        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self.video_processor = VideoProcessor(do_resize=True, vae_scale_factor=self.vae_scale_factor)
 
     def _encode_image(
         self,
@@ -142,12 +186,12 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
         device: Union[str, torch.device],
         num_videos_per_prompt: int,
         do_classifier_free_guidance: bool,
-    ) -> torch.FloatTensor:
+    ) -> torch.Tensor:
         dtype = next(self.image_encoder.parameters()).dtype
 
         if not isinstance(image, torch.Tensor):
-            image = self.image_processor.pil_to_numpy(image)
-            image = self.image_processor.numpy_to_pt(image)
+            image = self.video_processor.pil_to_numpy(image)
+            image = self.video_processor.numpy_to_pt(image)
 
             # We normalize the image before resizing to match with the original implementation.
             # Then we unnormalize it after resizing.
@@ -194,6 +238,9 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
         image = image.to(device=device)
         image_latents = self.vae.encode(image).latent_dist.mode()
 
+        # duplicate image_latents for each generation per prompt, using mps friendly method
+        image_latents = image_latents.repeat(num_videos_per_prompt, 1, 1, 1)
+
         if do_classifier_free_guidance:
             negative_image_latents = torch.zeros_like(image_latents)
 
@@ -202,9 +249,6 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
             # to avoid doing two forward passes
             image_latents = torch.cat([negative_image_latents, image_latents])
 
-        # duplicate image_latents for each generation per prompt, using mps friendly method
-        image_latents = image_latents.repeat(num_videos_per_prompt, 1, 1, 1)
-
         return image_latents
 
     def _get_add_time_ids(
@@ -235,7 +279,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
 
         return add_time_ids
 
-    def decode_latents(self, latents: torch.FloatTensor, num_frames: int, decode_chunk_size: int = 14):
+    def decode_latents(self, latents: torch.Tensor, num_frames: int, decode_chunk_size: int = 14):
         # [batch, frames, channels, height, width] -> [batch*frames, channels, height, width]
         latents = latents.flatten(0, 1)
 
@@ -271,7 +315,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
             and not isinstance(image, list)
         ):
             raise ValueError(
-                "`image` has to be of type `torch.FloatTensor` or `PIL.Image.Image` or `List[PIL.Image.Image]` but is"
+                "`image` has to be of type `torch.Tensor` or `PIL.Image.Image` or `List[PIL.Image.Image]` but is"
                 f" {type(image)}"
             )
 
@@ -288,7 +332,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
         dtype: torch.dtype,
         device: Union[str, torch.device],
         generator: torch.Generator,
-        latents: Optional[torch.FloatTensor] = None,
+        latents: Optional[torch.Tensor] = None,
     ):
         shape = (
             batch_size,
@@ -333,11 +377,12 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
     @replace_example_docstring(EXAMPLE_DOC_STRING)
     def __call__(
         self,
-        image: Union[PIL.Image.Image, List[PIL.Image.Image], torch.FloatTensor],
+        image: Union[PIL.Image.Image, List[PIL.Image.Image], torch.Tensor],
         height: int = 576,
         width: int = 1024,
         num_frames: Optional[int] = None,
         num_inference_steps: int = 25,
+        sigmas: Optional[List[float]] = None,
         min_guidance_scale: float = 1.0,
         max_guidance_scale: float = 3.0,
         fps: int = 7,
@@ -346,7 +391,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
         decode_chunk_size: Optional[int] = None,
         num_videos_per_prompt: Optional[int] = 1,
         generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
-        latents: Optional[torch.FloatTensor] = None,
+        latents: Optional[torch.Tensor] = None,
         output_type: Optional[str] = "pil",
         callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
         callback_on_step_end_tensor_inputs: List[str] = ["latents"],
@@ -356,39 +401,46 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
         The call function to the pipeline for generation.
 
         Args:
-            image (`PIL.Image.Image` or `List[PIL.Image.Image]` or `torch.FloatTensor`):
-                Image(s) to guide image generation. If you provide a tensor, the expected value range is between `[0, 1]`.
+            image (`PIL.Image.Image` or `List[PIL.Image.Image]` or `torch.Tensor`):
+                Image(s) to guide image generation. If you provide a tensor, the expected value range is between `[0,
+                1]`.
             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_frames (`int`, *optional*):
-                The number of video frames to generate. Defaults to `self.unet.config.num_frames`
-                (14 for `stable-video-diffusion-img2vid` and to 25 for `stable-video-diffusion-img2vid-xt`).
+                The number of video frames to generate. Defaults to `self.unet.config.num_frames` (14 for
+                `stable-video-diffusion-img2vid` and to 25 for `stable-video-diffusion-img2vid-xt`).
             num_inference_steps (`int`, *optional*, defaults to 25):
                 The number of denoising steps. More denoising steps usually lead to a higher quality video at the
                 expense of slower inference. This parameter is modulated by `strength`.
+            sigmas (`List[float]`, *optional*):
+                Custom sigmas to use for the denoising process with schedulers which support a `sigmas` argument in
+                their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is passed
+                will be used.
             min_guidance_scale (`float`, *optional*, defaults to 1.0):
                 The minimum guidance scale. Used for the classifier free guidance with first frame.
             max_guidance_scale (`float`, *optional*, defaults to 3.0):
                 The maximum guidance scale. Used for the classifier free guidance with last frame.
             fps (`int`, *optional*, defaults to 7):
-                Frames per second. The rate at which the generated images shall be exported to a video after generation.
-                Note that Stable Diffusion Video's UNet was micro-conditioned on fps-1 during training.
+                Frames per second. The rate at which the generated images shall be exported to a video after
+                generation. Note that Stable Diffusion Video's UNet was micro-conditioned on fps-1 during training.
             motion_bucket_id (`int`, *optional*, defaults to 127):
                 Used for conditioning the amount of motion for the generation. The higher the number the more motion
                 will be in the video.
             noise_aug_strength (`float`, *optional*, defaults to 0.02):
-                The amount of noise added to the init image, the higher it is the less the video will look like the init image. Increase it for more motion.
+                The amount of noise added to the init image, the higher it is the less the video will look like the
+                init image. Increase it for more motion.
             decode_chunk_size (`int`, *optional*):
-                The number of frames to decode at a time. Higher chunk size leads to better temporal consistency at the expense of more memory usage. By default, the decoder decodes all frames at once for maximal
-                quality. For lower memory usage, reduce `decode_chunk_size`.
+                The number of frames to decode at a time. Higher chunk size leads to better temporal consistency at the
+                expense of more memory usage. By default, the decoder decodes all frames at once for maximal quality.
+                For lower memory usage, reduce `decode_chunk_size`.
             num_videos_per_prompt (`int`, *optional*, defaults to 1):
                 The number of videos to generate per prompt.
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
                 generation deterministic.
-            latents (`torch.FloatTensor`, *optional*):
+            latents (`torch.Tensor`, *optional*):
                 Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for video
                 generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
                 tensor is generated by sampling using the supplied random `generator`.
@@ -398,7 +450,8 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
                 A function that is called at the end of each denoising step during inference. The function is called
                 with the following arguments:
                     `callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict)`.
-                `callback_kwargs` will include a list of all tensors as specified by `callback_on_step_end_tensor_inputs`.
+                `callback_kwargs` will include a list of all tensors as specified by
+                `callback_on_step_end_tensor_inputs`.
             callback_on_step_end_tensor_inputs (`List`, *optional*):
                 The list of tensor inputs for the `callback_on_step_end` function. The tensors specified in the list
                 will be passed as `callback_kwargs` argument. You will only be able to include variables listed in the
@@ -411,8 +464,9 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
 
         Returns:
             [`~pipelines.stable_diffusion.StableVideoDiffusionPipelineOutput`] or `tuple`:
-                If `return_dict` is `True`, [`~pipelines.stable_diffusion.StableVideoDiffusionPipelineOutput`] is returned,
-                otherwise a `tuple` of (`List[List[PIL.Image.Image]]` or `np.ndarray` or `torch.FloatTensor`) is returned.
+                If `return_dict` is `True`, [`~pipelines.stable_diffusion.StableVideoDiffusionPipelineOutput`] is
+                returned, otherwise a `tuple` of (`List[List[PIL.Image.Image]]` or `np.ndarray` or `torch.Tensor`) is
+                returned.
         """
         # 0. Default height and width to unet
         height = height or self.unet.config.sample_size * self.vae_scale_factor
@@ -445,7 +499,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
         fps = fps - 1
 
         # 4. Encode input image using VAE
-        image = self.image_processor.preprocess(image, height=height, width=width).to(device)
+        image = self.video_processor.preprocess(image, height=height, width=width).to(device)
         noise = randn_tensor(image.shape, generator=generator, device=device, dtype=image.dtype)
         image = image + noise_aug_strength * noise
 
@@ -482,8 +536,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
         added_time_ids = added_time_ids.to(device)
 
         # 6. Prepare timesteps
-        self.scheduler.set_timesteps(num_inference_steps, device=device)
-        timesteps = self.scheduler.timesteps
+        timesteps, num_inference_steps = retrieve_timesteps(self.scheduler, num_inference_steps, device, None, sigmas)
 
         # 7. Prepare latent variables
         num_channels_latents = self.unet.config.in_channels
@@ -552,7 +605,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
             if needs_upcasting:
                 self.vae.to(dtype=torch.float16)
             frames = self.decode_latents(latents, num_frames, decode_chunk_size)
-            frames = tensor2vid(frames, self.image_processor, output_type=output_type)
+            frames = self.video_processor.postprocess_video(video=frames, output_type=output_type)
         else:
             frames = latents
 
@@ -627,7 +680,7 @@ def _filter2d(input, kernel):
 
     height, width = tmp_kernel.shape[-2:]
 
-    padding_shape: list[int] = _compute_padding([height, width])
+    padding_shape: List[int] = _compute_padding([height, width])
     input = torch.nn.functional.pad(input, padding_shape, mode="reflect")
 
     # kernel and input tensor reshape to align element-wise or batch-wise params