Add files using upload-large-folder tool

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/deprecated/__init__.py

@@ -0,0 +1,153 @@
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    DIFFUSERS_SLOW_IMPORT,
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    get_objects_from_module,
+    is_librosa_available,
+    is_note_seq_available,
+    is_torch_available,
+    is_transformers_available,
+)
+
+
+_dummy_objects = {}
+_import_structure = {}
+
+try:
+    if not is_torch_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    from ...utils import dummy_pt_objects
+
+    _dummy_objects.update(get_objects_from_module(dummy_pt_objects))
+else:
+    _import_structure["latent_diffusion_uncond"] = ["LDMPipeline"]
+    _import_structure["pndm"] = ["PNDMPipeline"]
+    _import_structure["repaint"] = ["RePaintPipeline"]
+    _import_structure["score_sde_ve"] = ["ScoreSdeVePipeline"]
+    _import_structure["stochastic_karras_ve"] = ["KarrasVePipeline"]
+
+try:
+    if not (is_transformers_available() and is_torch_available()):
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    from ...utils import dummy_torch_and_transformers_objects
+
+    _dummy_objects.update(get_objects_from_module(dummy_torch_and_transformers_objects))
+else:
+    _import_structure["alt_diffusion"] = [
+        "AltDiffusionImg2ImgPipeline",
+        "AltDiffusionPipeline",
+        "AltDiffusionPipelineOutput",
+    ]
+    _import_structure["versatile_diffusion"] = [
+        "VersatileDiffusionDualGuidedPipeline",
+        "VersatileDiffusionImageVariationPipeline",
+        "VersatileDiffusionPipeline",
+        "VersatileDiffusionTextToImagePipeline",
+    ]
+    _import_structure["vq_diffusion"] = ["VQDiffusionPipeline"]
+    _import_structure["stable_diffusion_variants"] = [
+        "CycleDiffusionPipeline",
+        "StableDiffusionInpaintPipelineLegacy",
+        "StableDiffusionPix2PixZeroPipeline",
+        "StableDiffusionParadigmsPipeline",
+        "StableDiffusionModelEditingPipeline",
+    ]
+
+try:
+    if not (is_torch_available() and is_librosa_available()):
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    from ...utils import dummy_torch_and_librosa_objects  # noqa F403
+
+    _dummy_objects.update(get_objects_from_module(dummy_torch_and_librosa_objects))
+
+else:
+    _import_structure["audio_diffusion"] = ["AudioDiffusionPipeline", "Mel"]
+
+try:
+    if not (is_transformers_available() and is_torch_available() and is_note_seq_available()):
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    from ...utils import dummy_transformers_and_torch_and_note_seq_objects  # noqa F403
+
+    _dummy_objects.update(get_objects_from_module(dummy_transformers_and_torch_and_note_seq_objects))
+
+else:
+    _import_structure["spectrogram_diffusion"] = ["MidiProcessor", "SpectrogramDiffusionPipeline"]
+
+
+if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
+    try:
+        if not is_torch_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        from ...utils.dummy_pt_objects import *
+
+    else:
+        from .latent_diffusion_uncond import LDMPipeline
+        from .pndm import PNDMPipeline
+        from .repaint import RePaintPipeline
+        from .score_sde_ve import ScoreSdeVePipeline
+        from .stochastic_karras_ve import KarrasVePipeline
+
+    try:
+        if not (is_transformers_available() and is_torch_available()):
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        from ...utils.dummy_torch_and_transformers_objects import *
+
+    else:
+        from .alt_diffusion import AltDiffusionImg2ImgPipeline, AltDiffusionPipeline, AltDiffusionPipelineOutput
+        from .audio_diffusion import AudioDiffusionPipeline, Mel
+        from .spectrogram_diffusion import SpectrogramDiffusionPipeline
+        from .stable_diffusion_variants import (
+            CycleDiffusionPipeline,
+            StableDiffusionInpaintPipelineLegacy,
+            StableDiffusionModelEditingPipeline,
+            StableDiffusionParadigmsPipeline,
+            StableDiffusionPix2PixZeroPipeline,
+        )
+        from .stochastic_karras_ve import KarrasVePipeline
+        from .versatile_diffusion import (
+            VersatileDiffusionDualGuidedPipeline,
+            VersatileDiffusionImageVariationPipeline,
+            VersatileDiffusionPipeline,
+            VersatileDiffusionTextToImagePipeline,
+        )
+        from .vq_diffusion import VQDiffusionPipeline
+
+    try:
+        if not (is_torch_available() and is_librosa_available()):
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        from ...utils.dummy_torch_and_librosa_objects import *
+    else:
+        from .audio_diffusion import AudioDiffusionPipeline, Mel
+
+    try:
+        if not (is_transformers_available() and is_torch_available() and is_note_seq_available()):
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        from ...utils.dummy_transformers_and_torch_and_note_seq_objects import *  # noqa F403
+    else:
+        from .spectrogram_diffusion import (
+            MidiProcessor,
+            SpectrogramDiffusionPipeline,
+        )
+
+
+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)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/deprecated/alt_diffusion/__init__.py

@@ -0,0 +1,53 @@
+from typing import TYPE_CHECKING
+
+from ....utils import (
+    DIFFUSERS_SLOW_IMPORT,
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    get_objects_from_module,
+    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
+
+    _dummy_objects.update(get_objects_from_module(dummy_torch_and_transformers_objects))
+else:
+    _import_structure["modeling_roberta_series"] = ["RobertaSeriesModelWithTransformation"]
+    _import_structure["pipeline_alt_diffusion"] = ["AltDiffusionPipeline"]
+    _import_structure["pipeline_alt_diffusion_img2img"] = ["AltDiffusionImg2ImgPipeline"]
+
+    _import_structure["pipeline_output"] = ["AltDiffusionPipelineOutput"]
+
+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 .modeling_roberta_series import RobertaSeriesModelWithTransformation
+        from .pipeline_alt_diffusion import AltDiffusionPipeline
+        from .pipeline_alt_diffusion_img2img import AltDiffusionImg2ImgPipeline
+        from .pipeline_output import AltDiffusionPipelineOutput
+
+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)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/deprecated/alt_diffusion/modeling_roberta_series.py

@@ -0,0 +1,124 @@
+from dataclasses import dataclass
+from typing import Optional, Tuple
+
+import torch
+from torch import nn
+from transformers import RobertaPreTrainedModel, XLMRobertaConfig, XLMRobertaModel
+from transformers.utils import ModelOutput
+
+
+@dataclass
+class TransformationModelOutput(ModelOutput):
+    """
+    Base class for text model's outputs that also contains a pooling of the last hidden states.
+
+    Args:
+        text_embeds (`torch.Tensor` of shape `(batch_size, output_dim)` *optional* returned when model is initialized with `with_projection=True`):
+            The text embeddings obtained by applying the projection layer to the pooler_output.
+        last_hidden_state (`torch.Tensor` of shape `(batch_size, sequence_length, hidden_size)`):
+            Sequence of hidden-states at the output of the last layer of the model.
+        hidden_states (`tuple(torch.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
+            Tuple of `torch.Tensor` (one for the output of the embeddings, if the model has an embedding layer, + one
+            for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.
+
+            Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
+        attentions (`tuple(torch.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`):
+            Tuple of `torch.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
+            sequence_length)`.
+
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
+            heads.
+    """
+
+    projection_state: Optional[torch.Tensor] = None
+    last_hidden_state: torch.Tensor = None
+    hidden_states: Optional[Tuple[torch.Tensor]] = None
+    attentions: Optional[Tuple[torch.Tensor]] = None
+
+
+class RobertaSeriesConfig(XLMRobertaConfig):
+    def __init__(
+        self,
+        pad_token_id=1,
+        bos_token_id=0,
+        eos_token_id=2,
+        project_dim=512,
+        pooler_fn="cls",
+        learn_encoder=False,
+        use_attention_mask=True,
+        **kwargs,
+    ):
+        super().__init__(pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs)
+        self.project_dim = project_dim
+        self.pooler_fn = pooler_fn
+        self.learn_encoder = learn_encoder
+        self.use_attention_mask = use_attention_mask
+
+
+class RobertaSeriesModelWithTransformation(RobertaPreTrainedModel):
+    _keys_to_ignore_on_load_unexpected = [r"pooler", r"logit_scale"]
+    _keys_to_ignore_on_load_missing = [r"position_ids", r"predictions.decoder.bias"]
+    base_model_prefix = "roberta"
+    config_class = RobertaSeriesConfig
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.roberta = XLMRobertaModel(config)
+        self.transformation = nn.Linear(config.hidden_size, config.project_dim)
+        self.has_pre_transformation = getattr(config, "has_pre_transformation", False)
+        if self.has_pre_transformation:
+            self.transformation_pre = nn.Linear(config.hidden_size, config.project_dim)
+            self.pre_LN = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.post_init()
+
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        encoder_hidden_states: Optional[torch.Tensor] = None,
+        encoder_attention_mask: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+    ):
+        r""" """
+
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.base_model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_attention_mask,
+            output_attentions=output_attentions,
+            output_hidden_states=True if self.has_pre_transformation else output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        if self.has_pre_transformation:
+            sequence_output2 = outputs["hidden_states"][-2]
+            sequence_output2 = self.pre_LN(sequence_output2)
+            projection_state2 = self.transformation_pre(sequence_output2)
+
+            return TransformationModelOutput(
+                projection_state=projection_state2,
+                last_hidden_state=outputs.last_hidden_state,
+                hidden_states=outputs.hidden_states,
+                attentions=outputs.attentions,
+            )
+        else:
+            projection_state = self.transformation(outputs.last_hidden_state)
+            return TransformationModelOutput(
+                projection_state=projection_state,
+                last_hidden_state=outputs.last_hidden_state,
+                hidden_states=outputs.hidden_states,
+                attentions=outputs.attentions,
+            )

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/deprecated/alt_diffusion/pipeline_alt_diffusion.py

@@ -0,0 +1,990 @@
+# Copyright 2025 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import inspect
+from typing import Any, Callable, Dict, List, Optional, Union
+
+import torch
+from packaging import version
+from transformers import CLIPImageProcessor, CLIPVisionModelWithProjection, XLMRobertaTokenizer
+
+from ....configuration_utils import FrozenDict
+from ....image_processor import PipelineImageInput, VaeImageProcessor
+from ....loaders import (
+    FromSingleFileMixin,
+    IPAdapterMixin,
+    StableDiffusionLoraLoaderMixin,
+    TextualInversionLoaderMixin,
+)
+from ....models import AutoencoderKL, ImageProjection, UNet2DConditionModel
+from ....models.lora import adjust_lora_scale_text_encoder
+from ....schedulers import KarrasDiffusionSchedulers
+from ....utils import (
+    USE_PEFT_BACKEND,
+    deprecate,
+    logging,
+    replace_example_docstring,
+    scale_lora_layers,
+    unscale_lora_layers,
+)
+from ....utils.torch_utils import randn_tensor
+from ...pipeline_utils import DiffusionPipeline, StableDiffusionMixin
+from ...stable_diffusion.safety_checker import StableDiffusionSafetyChecker
+from .modeling_roberta_series import RobertaSeriesModelWithTransformation
+from .pipeline_output import AltDiffusionPipelineOutput
+
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> import torch
+        >>> from diffusers import AltDiffusionPipeline
+
+        >>> pipe = AltDiffusionPipeline.from_pretrained("BAAI/AltDiffusion-m9", torch_dtype=torch.float16)
+        >>> pipe = pipe.to("cuda")
+
+        >>> # "dark elf princess, highly detailed, d & d, fantasy, highly detailed, digital painting, trending on artstation, concept art, sharp focus, illustration, art by artgerm and greg rutkowski and fuji choko and viktoria gavrilenko and hoang lap"
+        >>> prompt = "黑暗精灵公主，非常详细，幻想，非常详细，数字绘画，概念艺术，敏锐的焦点，插图"
+        >>> image = pipe(prompt).images[0]
+        ```
+"""
+
+
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.rescale_noise_cfg
+def rescale_noise_cfg(noise_cfg, noise_pred_text, guidance_rescale=0.0):
+    r"""
+    Rescales `noise_cfg` tensor based on `guidance_rescale` to improve image quality and fix overexposure. Based on
+    Section 3.4 from [Common Diffusion Noise Schedules and Sample Steps are
+    Flawed](https://huggingface.co/papers/2305.08891).
+
+    Args:
+        noise_cfg (`torch.Tensor`):
+            The predicted noise tensor for the guided diffusion process.
+        noise_pred_text (`torch.Tensor`):
+            The predicted noise tensor for the text-guided diffusion process.
+        guidance_rescale (`float`, *optional*, defaults to 0.0):
+            A rescale factor applied to the noise predictions.
+
+    Returns:
+        noise_cfg (`torch.Tensor`): The rescaled noise prediction tensor.
+    """
+    std_text = noise_pred_text.std(dim=list(range(1, noise_pred_text.ndim)), keepdim=True)
+    std_cfg = noise_cfg.std(dim=list(range(1, noise_cfg.ndim)), keepdim=True)
+    # rescale the results from guidance (fixes overexposure)
+    noise_pred_rescaled = noise_cfg * (std_text / std_cfg)
+    # mix with the original results from guidance by factor guidance_rescale to avoid "plain looking" images
+    noise_cfg = guidance_rescale * noise_pred_rescaled + (1 - guidance_rescale) * noise_cfg
+    return noise_cfg
+
+
+# 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`.
+
+    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
+
+
+class AltDiffusionPipeline(
+    DiffusionPipeline,
+    StableDiffusionMixin,
+    TextualInversionLoaderMixin,
+    StableDiffusionLoraLoaderMixin,
+    IPAdapterMixin,
+    FromSingleFileMixin,
+):
+    r"""
+    Pipeline for text-to-image generation using Alt Diffusion.
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods
+    implemented for all pipelines (downloading, saving, running on a particular device, etc.).
+
+    The pipeline also inherits the following loading methods:
+        - [`~loaders.TextualInversionLoaderMixin.load_textual_inversion`] for loading textual inversion embeddings
+        - [`~loaders.StableDiffusionLoraLoaderMixin.load_lora_weights`] for loading LoRA weights
+        - [`~loaders.StableDiffusionLoraLoaderMixin.save_lora_weights`] for saving LoRA weights
+        - [`~loaders.FromSingleFileMixin.from_single_file`] for loading `.ckpt` files
+        - [`~loaders.IPAdapterMixin.load_ip_adapter`] for loading IP Adapters
+
+    Args:
+        vae ([`AutoencoderKL`]):
+            Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations.
+        text_encoder ([`~transformers.RobertaSeriesModelWithTransformation`]):
+            Frozen text-encoder ([clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14)).
+        tokenizer ([`~transformers.XLMRobertaTokenizer`]):
+            A `XLMRobertaTokenizer` to tokenize text.
+        unet ([`UNet2DConditionModel`]):
+            A `UNet2DConditionModel` to denoise the encoded image latents.
+        scheduler ([`SchedulerMixin`]):
+            A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of
+            [`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`].
+        safety_checker ([`StableDiffusionSafetyChecker`]):
+            Classification module that estimates whether generated images could be considered offensive or harmful.
+            Please refer to the [model card](https://huggingface.co/runwayml/stable-diffusion-v1-5) for more details
+            about a model's potential harms.
+        feature_extractor ([`~transformers.CLIPImageProcessor`]):
+            A `CLIPImageProcessor` to extract features from generated images; used as inputs to the `safety_checker`.
+    """
+
+    model_cpu_offload_seq = "text_encoder->image_encoder->unet->vae"
+    _optional_components = ["safety_checker", "feature_extractor", "image_encoder"]
+    _exclude_from_cpu_offload = ["safety_checker"]
+    _callback_tensor_inputs = ["latents", "prompt_embeds", "negative_prompt_embeds"]
+
+    def __init__(
+        self,
+        vae: AutoencoderKL,
+        text_encoder: RobertaSeriesModelWithTransformation,
+        tokenizer: XLMRobertaTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: KarrasDiffusionSchedulers,
+        safety_checker: StableDiffusionSafetyChecker,
+        feature_extractor: CLIPImageProcessor,
+        image_encoder: CLIPVisionModelWithProjection = None,
+        requires_safety_checker: bool = True,
+    ):
+        super().__init__()
+
+        if scheduler is not None and getattr(scheduler.config, "steps_offset", 1) != 1:
+            deprecation_message = (
+                f"The configuration file of this scheduler: {scheduler} is outdated. `steps_offset`"
+                f" should be set to 1 instead of {scheduler.config.steps_offset}. Please make sure "
+                "to update the config accordingly as leaving `steps_offset` might led to incorrect results"
+                " in future versions. If you have downloaded this checkpoint from the Hugging Face Hub,"
+                " it would be very nice if you could open a Pull request for the `scheduler/scheduler_config.json`"
+                " file"
+            )
+            deprecate("steps_offset!=1", "1.0.0", deprecation_message, standard_warn=False)
+            new_config = dict(scheduler.config)
+            new_config["steps_offset"] = 1
+            scheduler._internal_dict = FrozenDict(new_config)
+
+        if scheduler is not None and getattr(scheduler.config, "clip_sample", False) is True:
+            deprecation_message = (
+                f"The configuration file of this scheduler: {scheduler} has not set the configuration `clip_sample`."
+                " `clip_sample` should be set to False in the configuration file. Please make sure to update the"
+                " config accordingly as not setting `clip_sample` in the config might lead to incorrect results in"
+                " future versions. If you have downloaded this checkpoint from the Hugging Face Hub, it would be very"
+                " nice if you could open a Pull request for the `scheduler/scheduler_config.json` file"
+            )
+            deprecate("clip_sample not set", "1.0.0", deprecation_message, standard_warn=False)
+            new_config = dict(scheduler.config)
+            new_config["clip_sample"] = False
+            scheduler._internal_dict = FrozenDict(new_config)
+
+        if safety_checker is None and requires_safety_checker:
+            logger.warning(
+                f"You have disabled the safety checker for {self.__class__} by passing `safety_checker=None`. Ensure"
+                " that you abide to the conditions of the Alt Diffusion license and do not expose unfiltered"
+                " results in services or applications open to the public. Both the diffusers team and Hugging Face"
+                " strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling"
+                " it only for use-cases that involve analyzing network behavior or auditing its results. For more"
+                " information, please have a look at https://github.com/huggingface/diffusers/pull/254 ."
+            )
+
+        if safety_checker is not None and feature_extractor is None:
+            raise ValueError(
+                "Make sure to define a feature extractor when loading {self.__class__} if you want to use the safety"
+                " checker. If you do not want to use the safety checker, you can pass `'safety_checker=None'` instead."
+            )
+
+        is_unet_version_less_0_9_0 = (
+            unet is not None
+            and hasattr(unet.config, "_diffusers_version")
+            and version.parse(version.parse(unet.config._diffusers_version).base_version) < version.parse("0.9.0.dev0")
+        )
+        is_unet_sample_size_less_64 = (
+            unet is not None and hasattr(unet.config, "sample_size") and unet.config.sample_size < 64
+        )
+        if is_unet_version_less_0_9_0 and is_unet_sample_size_less_64:
+            deprecation_message = (
+                "The configuration file of the unet has set the default `sample_size` to smaller than"
+                " 64 which seems highly unlikely. If your checkpoint is a fine-tuned version of any of the"
+                " following: \n- CompVis/stable-diffusion-v1-4 \n- CompVis/stable-diffusion-v1-3 \n-"
+                " CompVis/stable-diffusion-v1-2 \n- CompVis/stable-diffusion-v1-1 \n- runwayml/stable-diffusion-v1-5"
+                " \n- runwayml/stable-diffusion-inpainting \n you should change 'sample_size' to 64 in the"
+                " configuration file. Please make sure to update the config accordingly as leaving `sample_size=32`"
+                " in the config might lead to incorrect results in future versions. If you have downloaded this"
+                " checkpoint from the Hugging Face Hub, it would be very nice if you could open a Pull request for"
+                " the `unet/config.json` file"
+            )
+            deprecate("sample_size<64", "1.0.0", deprecation_message, standard_warn=False)
+            new_config = dict(unet.config)
+            new_config["sample_size"] = 64
+            unet._internal_dict = FrozenDict(new_config)
+
+        self.register_modules(
+            vae=vae,
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            safety_checker=safety_checker,
+            feature_extractor=feature_extractor,
+            image_encoder=image_encoder,
+        )
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1) if getattr(self, "vae", None) else 8
+        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self.register_to_config(requires_safety_checker=requires_safety_checker)
+
+    def _encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        lora_scale: Optional[float] = None,
+        **kwargs,
+    ):
+        deprecation_message = "`_encode_prompt()` is deprecated and it will be removed in a future version. Use `encode_prompt()` instead. Also, be aware that the output format changed from a concatenated tensor to a tuple."
+        deprecate("_encode_prompt()", "1.0.0", deprecation_message, standard_warn=False)
+
+        prompt_embeds_tuple = self.encode_prompt(
+            prompt=prompt,
+            device=device,
+            num_images_per_prompt=num_images_per_prompt,
+            do_classifier_free_guidance=do_classifier_free_guidance,
+            negative_prompt=negative_prompt,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
+            lora_scale=lora_scale,
+            **kwargs,
+        )
+
+        # concatenate for backwards comp
+        prompt_embeds = torch.cat([prompt_embeds_tuple[1], prompt_embeds_tuple[0]])
+
+        return prompt_embeds
+
+    def encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        lora_scale: Optional[float] = None,
+        clip_skip: Optional[int] = None,
+    ):
+        r"""
+        Encodes the prompt into text encoder hidden states.
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                prompt to be encoded
+            device: (`torch.device`):
+                torch device
+            num_images_per_prompt (`int`):
+                number of images that should be generated per prompt
+            do_classifier_free_guidance (`bool`):
+                whether to use classifier free guidance or not
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. If not defined, one has to pass
+                `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
+                less than `1`).
+            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.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.
+            lora_scale (`float`, *optional*):
+                A LoRA scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded.
+            clip_skip (`int`, *optional*):
+                Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
+                the output of the pre-final layer will be used for computing the prompt embeddings.
+        """
+        # set lora scale so that monkey patched LoRA
+        # function of text encoder can correctly access it
+        if lora_scale is not None and isinstance(self, StableDiffusionLoraLoaderMixin):
+            self._lora_scale = lora_scale
+
+            # dynamically adjust the LoRA scale
+            if not USE_PEFT_BACKEND:
+                adjust_lora_scale_text_encoder(self.text_encoder, lora_scale)
+            else:
+                scale_lora_layers(self.text_encoder, lora_scale)
+
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        if prompt_embeds is None:
+            # textual inversion: process multi-vector tokens if necessary
+            if isinstance(self, TextualInversionLoaderMixin):
+                prompt = self.maybe_convert_prompt(prompt, self.tokenizer)
+
+            text_inputs = self.tokenizer(
+                prompt,
+                padding="max_length",
+                max_length=self.tokenizer.model_max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+            text_input_ids = text_inputs.input_ids
+            untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+            if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(
+                text_input_ids, untruncated_ids
+            ):
+                removed_text = self.tokenizer.batch_decode(
+                    untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1]
+                )
+                logger.warning(
+                    "The following part of your input was truncated because CLIP can only handle sequences up to"
+                    f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+                )
+
+            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+                attention_mask = text_inputs.attention_mask.to(device)
+            else:
+                attention_mask = None
+
+            if clip_skip is None:
+                prompt_embeds = self.text_encoder(text_input_ids.to(device), attention_mask=attention_mask)
+                prompt_embeds = prompt_embeds[0]
+            else:
+                prompt_embeds = self.text_encoder(
+                    text_input_ids.to(device), attention_mask=attention_mask, output_hidden_states=True
+                )
+                # Access the `hidden_states` first, that contains a tuple of
+                # all the hidden states from the encoder layers. Then index into
+                # the tuple to access the hidden states from the desired layer.
+                prompt_embeds = prompt_embeds[-1][-(clip_skip + 1)]
+                # We also need to apply the final LayerNorm here to not mess with the
+                # representations. The `last_hidden_states` that we typically use for
+                # obtaining the final prompt representations passes through the LayerNorm
+                # layer.
+                prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)
+
+        if self.text_encoder is not None:
+            prompt_embeds_dtype = self.text_encoder.dtype
+        elif self.unet is not None:
+            prompt_embeds_dtype = self.unet.dtype
+        else:
+            prompt_embeds_dtype = prompt_embeds.dtype
+
+        prompt_embeds = prompt_embeds.to(dtype=prompt_embeds_dtype, device=device)
+
+        bs_embed, seq_len, _ = prompt_embeds.shape
+        # duplicate text embeddings for each generation per prompt, using mps friendly method
+        prompt_embeds = prompt_embeds.repeat(1, num_images_per_prompt, 1)
+        prompt_embeds = prompt_embeds.view(bs_embed * num_images_per_prompt, seq_len, -1)
+
+        # get unconditional embeddings for classifier free guidance
+        if do_classifier_free_guidance and negative_prompt_embeds is None:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif prompt is not None and type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            # textual inversion: process multi-vector tokens if necessary
+            if isinstance(self, TextualInversionLoaderMixin):
+                uncond_tokens = self.maybe_convert_prompt(uncond_tokens, self.tokenizer)
+
+            max_length = prompt_embeds.shape[1]
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+
+            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+                attention_mask = uncond_input.attention_mask.to(device)
+            else:
+                attention_mask = None
+
+            negative_prompt_embeds = self.text_encoder(
+                uncond_input.input_ids.to(device),
+                attention_mask=attention_mask,
+            )
+            negative_prompt_embeds = negative_prompt_embeds[0]
+
+        if do_classifier_free_guidance:
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+            seq_len = negative_prompt_embeds.shape[1]
+
+            negative_prompt_embeds = negative_prompt_embeds.to(dtype=prompt_embeds_dtype, device=device)
+
+            negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt, 1)
+            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)
+
+        if isinstance(self, StableDiffusionLoraLoaderMixin) 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
+
+    def encode_image(self, image, device, num_images_per_prompt, output_hidden_states=None):
+        dtype = next(self.image_encoder.parameters()).dtype
+
+        if not isinstance(image, torch.Tensor):
+            image = self.feature_extractor(image, return_tensors="pt").pixel_values
+
+        image = image.to(device=device, dtype=dtype)
+        if output_hidden_states:
+            image_enc_hidden_states = self.image_encoder(image, output_hidden_states=True).hidden_states[-2]
+            image_enc_hidden_states = image_enc_hidden_states.repeat_interleave(num_images_per_prompt, dim=0)
+            uncond_image_enc_hidden_states = self.image_encoder(
+                torch.zeros_like(image), output_hidden_states=True
+            ).hidden_states[-2]
+            uncond_image_enc_hidden_states = uncond_image_enc_hidden_states.repeat_interleave(
+                num_images_per_prompt, dim=0
+            )
+            return image_enc_hidden_states, uncond_image_enc_hidden_states
+        else:
+            image_embeds = self.image_encoder(image).image_embeds
+            image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+            uncond_image_embeds = torch.zeros_like(image_embeds)
+
+            return image_embeds, uncond_image_embeds
+
+    def run_safety_checker(self, image, device, dtype):
+        if self.safety_checker is None:
+            has_nsfw_concept = None
+        else:
+            if torch.is_tensor(image):
+                feature_extractor_input = self.image_processor.postprocess(image, output_type="pil")
+            else:
+                feature_extractor_input = self.image_processor.numpy_to_pil(image)
+            safety_checker_input = self.feature_extractor(feature_extractor_input, return_tensors="pt").to(device)
+            image, has_nsfw_concept = self.safety_checker(
+                images=image, clip_input=safety_checker_input.pixel_values.to(dtype)
+            )
+        return image, has_nsfw_concept
+
+    def decode_latents(self, latents):
+        deprecation_message = "The decode_latents method is deprecated and will be removed in 1.0.0. Please use VaeImageProcessor.postprocess(...) instead"
+        deprecate("decode_latents", "1.0.0", deprecation_message, standard_warn=False)
+
+        latents = 1 / self.vae.config.scaling_factor * latents
+        image = self.vae.decode(latents, return_dict=False)[0]
+        image = (image / 2 + 0.5).clamp(0, 1)
+        # we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
+        image = image.cpu().permute(0, 2, 3, 1).float().numpy()
+        return image
+
+    def prepare_extra_step_kwargs(self, generator, eta):
+        # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
+        # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
+        # eta corresponds to η in DDIM paper: https://huggingface.co/papers/2010.02502
+        # and should be between [0, 1]
+
+        accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        extra_step_kwargs = {}
+        if accepts_eta:
+            extra_step_kwargs["eta"] = eta
+
+        # check if the scheduler accepts generator
+        accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        if accepts_generator:
+            extra_step_kwargs["generator"] = generator
+        return extra_step_kwargs
+
+    def check_inputs(
+        self,
+        prompt,
+        height,
+        width,
+        callback_steps,
+        negative_prompt=None,
+        prompt_embeds=None,
+        negative_prompt_embeds=None,
+        callback_on_step_end_tensor_inputs=None,
+    ):
+        if height % 8 != 0 or width % 8 != 0:
+            raise ValueError(f"`height` and `width` have to be divisible by 8 but are {height} and {width}.")
+
+        if callback_steps is not None and (not isinstance(callback_steps, int) or callback_steps <= 0):
+            raise ValueError(
+                f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
+                f" {type(callback_steps)}."
+            )
+        if callback_on_step_end_tensor_inputs is not None and not all(
+            k in self._callback_tensor_inputs for k in callback_on_step_end_tensor_inputs
+        ):
+            raise ValueError(
+                f"`callback_on_step_end_tensor_inputs` has to be in {self._callback_tensor_inputs}, but found {[k for k in callback_on_step_end_tensor_inputs if k not in self._callback_tensor_inputs]}"
+            )
+
+        if prompt is not None and prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
+                " only forward one of the two."
+            )
+        elif prompt is None and prompt_embeds is None:
+            raise ValueError(
+                "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
+            )
+        elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        if negative_prompt is not None and negative_prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
+                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
+            )
+
+        if prompt_embeds is not None and negative_prompt_embeds is not None:
+            if prompt_embeds.shape != negative_prompt_embeds.shape:
+                raise ValueError(
+                    "`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
+                    f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
+                    f" {negative_prompt_embeds.shape}."
+                )
+
+    def prepare_latents(self, batch_size, num_channels_latents, height, width, dtype, device, generator, latents=None):
+        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"
+                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+            )
+
+        if latents is None:
+            latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        else:
+            latents = latents.to(device)
+
+        # scale the initial noise by the standard deviation required by the scheduler
+        latents = latents * self.scheduler.init_noise_sigma
+        return latents
+
+    def get_guidance_scale_embedding(self, w, embedding_dim=512, dtype=torch.float32):
+        """
+        See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
+
+        Args:
+            timesteps (`torch.Tensor`):
+                generate embedding vectors at these timesteps
+            embedding_dim (`int`, *optional*, defaults to 512):
+                dimension of the embeddings to generate
+            dtype:
+                data type of the generated embeddings
+
+        Returns:
+            `torch.Tensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
+        """
+        assert len(w.shape) == 1
+        w = w * 1000.0
+
+        half_dim = embedding_dim // 2
+        emb = torch.log(torch.tensor(10000.0)) / (half_dim - 1)
+        emb = torch.exp(torch.arange(half_dim, dtype=dtype) * -emb)
+        emb = w.to(dtype)[:, None] * emb[None, :]
+        emb = torch.cat([torch.sin(emb), torch.cos(emb)], dim=1)
+        if embedding_dim % 2 == 1:  # zero pad
+            emb = torch.nn.functional.pad(emb, (0, 1))
+        assert emb.shape == (w.shape[0], embedding_dim)
+        return emb
+
+    @property
+    def guidance_scale(self):
+        return self._guidance_scale
+
+    @property
+    def guidance_rescale(self):
+        return self._guidance_rescale
+
+    @property
+    def clip_skip(self):
+        return self._clip_skip
+
+    # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
+    # of the Imagen paper: https://huggingface.co/papers/2205.11487 . `guidance_scale = 1`
+    # corresponds to doing no classifier free guidance.
+    @property
+    def do_classifier_free_guidance(self):
+        return self._guidance_scale > 1 and self.unet.config.time_cond_proj_dim is None
+
+    @property
+    def cross_attention_kwargs(self):
+        return self._cross_attention_kwargs
+
+    @property
+    def num_timesteps(self):
+        return self._num_timesteps
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]] = None,
+        height: Optional[int] = None,
+        width: Optional[int] = None,
+        num_inference_steps: int = 50,
+        timesteps: List[int] = None,
+        sigmas: List[float] = None,
+        guidance_scale: float = 7.5,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        num_images_per_prompt: Optional[int] = 1,
+        eta: float = 0.0,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        ip_adapter_image: Optional[PipelineImageInput] = None,
+        output_type: Optional[str] = "pil",
+        return_dict: bool = True,
+        cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+        guidance_rescale: float = 0.0,
+        clip_skip: Optional[int] = None,
+        callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
+        callback_on_step_end_tensor_inputs: List[str] = ["latents"],
+        **kwargs,
+    ):
+        r"""
+        The call function to the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
+            height (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`):
+                The width in pixels of the generated image.
+            num_inference_steps (`int`, *optional*, defaults to 50):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            timesteps (`List[int]`, *optional*):
+                Custom timesteps to use for the denoising process with schedulers which support a `timesteps` argument
+                in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is
+                passed will be used. Must be in descending order.
+            guidance_scale (`float`, *optional*, defaults to 7.5):
+                A higher guidance scale value encourages the model to generate images closely linked to the text
+                `prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide what to not include in image generation. If not defined, you need to
+                pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`).
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            eta (`float`, *optional*, defaults to 0.0):
+                Corresponds to parameter eta (η) from the [DDIM](https://huggingface.co/papers/2010.02502) paper. Only
+                applies to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers.
+            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.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 is generated by sampling using the supplied random `generator`.
+            prompt_embeds (`torch.Tensor`, *optional*):
+                Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
+                provided, text embeddings are generated from the `prompt` input argument.
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
+                Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
+                not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
+            ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generated image. Choose between `PIL.Image` or `np.array`.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.stable_diffusion.AltDiffusionPipelineOutput`] instead of a
+                plain tuple.
+            cross_attention_kwargs (`dict`, *optional*):
+                A kwargs dictionary that if specified is passed along to the [`AttentionProcessor`] as defined in
+                [`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+            guidance_rescale (`float`, *optional*, defaults to 0.0):
+                Guidance rescale factor from [Common Diffusion Noise Schedules and Sample Steps are
+                Flawed](https://huggingface.co/papers/2305.08891). Guidance rescale factor should fix overexposure when
+                using zero terminal SNR.
+            clip_skip (`int`, *optional*):
+                Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
+                the output of the pre-final layer will be used for computing the prompt embeddings.
+            callback_on_step_end (`Callable`, *optional*):
+                A function that calls at the end of each denoising steps during the 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_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
+                `._callback_tensor_inputs` attribute of your pipeline class.
+
+        Examples:
+
+        Returns:
+            [`~pipelines.stable_diffusion.AltDiffusionPipelineOutput`] or `tuple`:
+                If `return_dict` is `True`, [`~pipelines.stable_diffusion.AltDiffusionPipelineOutput`] is returned,
+                otherwise a `tuple` is returned where the first element is a list with the generated images and the
+                second element is a list of `bool`s indicating whether the corresponding generated image contains
+                "not-safe-for-work" (nsfw) content.
+        """
+
+        callback = kwargs.pop("callback", None)
+        callback_steps = kwargs.pop("callback_steps", None)
+
+        if callback is not None:
+            deprecate(
+                "callback",
+                "1.0.0",
+                "Passing `callback` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`",
+            )
+        if callback_steps is not None:
+            deprecate(
+                "callback_steps",
+                "1.0.0",
+                "Passing `callback_steps` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`",
+            )
+
+        # 0. Default height and width to unet
+        height = height or self.unet.config.sample_size * self.vae_scale_factor
+        width = width or self.unet.config.sample_size * self.vae_scale_factor
+        # to deal with lora scaling and other possible forward hooks
+
+        # 1. Check inputs. Raise error if not correct
+        self.check_inputs(
+            prompt,
+            height,
+            width,
+            callback_steps,
+            negative_prompt,
+            prompt_embeds,
+            negative_prompt_embeds,
+            callback_on_step_end_tensor_inputs,
+        )
+
+        self._guidance_scale = guidance_scale
+        self._guidance_rescale = guidance_rescale
+        self._clip_skip = clip_skip
+        self._cross_attention_kwargs = cross_attention_kwargs
+
+        # 2. Define call parameters
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        device = self._execution_device
+
+        # 3. Encode input prompt
+        lora_scale = (
+            self.cross_attention_kwargs.get("scale", None) if self.cross_attention_kwargs is not None else None
+        )
+
+        prompt_embeds, negative_prompt_embeds = self.encode_prompt(
+            prompt,
+            device,
+            num_images_per_prompt,
+            self.do_classifier_free_guidance,
+            negative_prompt,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
+            lora_scale=lora_scale,
+            clip_skip=self.clip_skip,
+        )
+
+        # For classifier free guidance, we need to do two forward passes.
+        # Here we concatenate the unconditional and text embeddings into a single batch
+        # to avoid doing two forward passes
+        if self.do_classifier_free_guidance:
+            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
+
+        if ip_adapter_image is not None:
+            output_hidden_state = False if isinstance(self.unet.encoder_hid_proj, ImageProjection) else True
+            image_embeds, negative_image_embeds = self.encode_image(
+                ip_adapter_image, device, num_images_per_prompt, output_hidden_state
+            )
+            if self.do_classifier_free_guidance:
+                image_embeds = torch.cat([negative_image_embeds, image_embeds])
+
+        # 4. Prepare timesteps
+        timesteps, num_inference_steps = retrieve_timesteps(
+            self.scheduler, num_inference_steps, device, timesteps, sigmas
+        )
+
+        # 5. Prepare latent variables
+        num_channels_latents = self.unet.config.in_channels
+        latents = self.prepare_latents(
+            batch_size * num_images_per_prompt,
+            num_channels_latents,
+            height,
+            width,
+            prompt_embeds.dtype,
+            device,
+            generator,
+            latents,
+        )
+
+        # 6. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
+        extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)
+
+        # 6.1 Add image embeds for IP-Adapter
+        added_cond_kwargs = {"image_embeds": image_embeds} if ip_adapter_image is not None else None
+
+        # 6.2 Optionally get Guidance Scale Embedding
+        timestep_cond = None
+        if self.unet.config.time_cond_proj_dim is not None:
+            guidance_scale_tensor = torch.tensor(self.guidance_scale - 1).repeat(batch_size * num_images_per_prompt)
+            timestep_cond = self.get_guidance_scale_embedding(
+                guidance_scale_tensor, embedding_dim=self.unet.config.time_cond_proj_dim
+            ).to(device=device, dtype=latents.dtype)
+
+        # 7. Denoising loop
+        num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order
+        self._num_timesteps = len(timesteps)
+        with self.progress_bar(total=num_inference_steps) as progress_bar:
+            for i, t in enumerate(timesteps):
+                # expand the latents if we are doing classifier free guidance
+                latent_model_input = torch.cat([latents] * 2) if self.do_classifier_free_guidance else latents
+                latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
+
+                # predict the noise residual
+                noise_pred = self.unet(
+                    latent_model_input,
+                    t,
+                    encoder_hidden_states=prompt_embeds,
+                    timestep_cond=timestep_cond,
+                    cross_attention_kwargs=self.cross_attention_kwargs,
+                    added_cond_kwargs=added_cond_kwargs,
+                    return_dict=False,
+                )[0]
+
+                # perform guidance
+                if self.do_classifier_free_guidance:
+                    noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
+                    noise_pred = noise_pred_uncond + self.guidance_scale * (noise_pred_text - noise_pred_uncond)
+
+                if self.do_classifier_free_guidance and self.guidance_rescale > 0.0:
+                    # Based on 3.4. in https://huggingface.co/papers/2305.08891
+                    noise_pred = rescale_noise_cfg(noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale)
+
+                # compute the previous noisy sample x_t -> x_t-1
+                latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs, return_dict=False)[0]
+
+                if callback_on_step_end is not None:
+                    callback_kwargs = {}
+                    for k in callback_on_step_end_tensor_inputs:
+                        callback_kwargs[k] = locals()[k]
+                    callback_outputs = callback_on_step_end(self, i, t, callback_kwargs)
+
+                    latents = callback_outputs.pop("latents", latents)
+                    prompt_embeds = callback_outputs.pop("prompt_embeds", prompt_embeds)
+                    negative_prompt_embeds = callback_outputs.pop("negative_prompt_embeds", negative_prompt_embeds)
+
+                # call the callback, if provided
+                if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
+                    progress_bar.update()
+                    if callback is not None and i % callback_steps == 0:
+                        step_idx = i // getattr(self.scheduler, "order", 1)
+                        callback(step_idx, t, latents)
+
+        if not output_type == "latent":
+            image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False, generator=generator)[
+                0
+            ]
+            image, has_nsfw_concept = self.run_safety_checker(image, device, prompt_embeds.dtype)
+        else:
+            image = latents
+            has_nsfw_concept = None
+
+        if has_nsfw_concept is None:
+            do_denormalize = [True] * image.shape[0]
+        else:
+            do_denormalize = [not has_nsfw for has_nsfw in has_nsfw_concept]
+
+        image = self.image_processor.postprocess(image, output_type=output_type, do_denormalize=do_denormalize)
+
+        # Offload all models
+        self.maybe_free_model_hooks()
+
+        if not return_dict:
+            return (image, has_nsfw_concept)
+
+        return AltDiffusionPipelineOutput(images=image, nsfw_content_detected=has_nsfw_concept)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/deprecated/alt_diffusion/pipeline_alt_diffusion_img2img.py

@@ -0,0 +1,1045 @@
+# Copyright 2025 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import inspect
+from typing import Any, Callable, Dict, List, Optional, Union
+
+import numpy as np
+import PIL.Image
+import torch
+from packaging import version
+from transformers import CLIPImageProcessor, CLIPVisionModelWithProjection, XLMRobertaTokenizer
+
+from ....configuration_utils import FrozenDict
+from ....image_processor import PipelineImageInput, VaeImageProcessor
+from ....loaders import (
+    FromSingleFileMixin,
+    IPAdapterMixin,
+    StableDiffusionLoraLoaderMixin,
+    TextualInversionLoaderMixin,
+)
+from ....models import AutoencoderKL, ImageProjection, UNet2DConditionModel
+from ....models.lora import adjust_lora_scale_text_encoder
+from ....schedulers import KarrasDiffusionSchedulers
+from ....utils import (
+    PIL_INTERPOLATION,
+    USE_PEFT_BACKEND,
+    deprecate,
+    logging,
+    replace_example_docstring,
+    scale_lora_layers,
+    unscale_lora_layers,
+)
+from ....utils.torch_utils import randn_tensor
+from ...pipeline_utils import DiffusionPipeline, StableDiffusionMixin
+from ...stable_diffusion.safety_checker import StableDiffusionSafetyChecker
+from .modeling_roberta_series import RobertaSeriesModelWithTransformation
+from .pipeline_output import AltDiffusionPipelineOutput
+
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> import requests
+        >>> import torch
+        >>> from PIL import Image
+        >>> from io import BytesIO
+
+        >>> from diffusers import AltDiffusionImg2ImgPipeline
+
+        >>> device = "cuda"
+        >>> model_id_or_path = "BAAI/AltDiffusion-m9"
+        >>> pipe = AltDiffusionImg2ImgPipeline.from_pretrained(model_id_or_path, torch_dtype=torch.float16)
+        >>> pipe = pipe.to(device)
+
+        >>> url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
+
+        >>> response = requests.get(url)
+        >>> init_image = Image.open(BytesIO(response.content)).convert("RGB")
+        >>> init_image = init_image.resize((768, 512))
+
+        >>> # "A fantasy landscape, trending on artstation"
+        >>> prompt = "幻想风景, artstation"
+
+        >>> images = pipe(prompt=prompt, image=init_image, strength=0.75, guidance_scale=7.5).images
+        >>> images[0].save("幻想风景.png")
+        ```
+"""
+
+
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_img2img.retrieve_latents
+def retrieve_latents(
+    encoder_output: torch.Tensor, generator: Optional[torch.Generator] = None, sample_mode: str = "sample"
+):
+    if hasattr(encoder_output, "latent_dist") and sample_mode == "sample":
+        return encoder_output.latent_dist.sample(generator)
+    elif hasattr(encoder_output, "latent_dist") and sample_mode == "argmax":
+        return encoder_output.latent_dist.mode()
+    elif hasattr(encoder_output, "latents"):
+        return encoder_output.latents
+    else:
+        raise AttributeError("Could not access latents of provided encoder_output")
+
+
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_img2img.preprocess
+def preprocess(image):
+    deprecation_message = "The preprocess method is deprecated and will be removed in diffusers 1.0.0. Please use VaeImageProcessor.preprocess(...) instead"
+    deprecate("preprocess", "1.0.0", deprecation_message, standard_warn=False)
+    if isinstance(image, torch.Tensor):
+        return image
+    elif isinstance(image, PIL.Image.Image):
+        image = [image]
+
+    if isinstance(image[0], PIL.Image.Image):
+        w, h = image[0].size
+        w, h = (x - x % 8 for x in (w, h))  # resize to integer multiple of 8
+
+        image = [np.array(i.resize((w, h), resample=PIL_INTERPOLATION["lanczos"]))[None, :] for i in image]
+        image = np.concatenate(image, axis=0)
+        image = np.array(image).astype(np.float32) / 255.0
+        image = image.transpose(0, 3, 1, 2)
+        image = 2.0 * image - 1.0
+        image = torch.from_numpy(image)
+    elif isinstance(image[0], torch.Tensor):
+        image = torch.cat(image, dim=0)
+    return image
+
+
+# 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`.
+
+    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
+
+
+class AltDiffusionImg2ImgPipeline(
+    DiffusionPipeline,
+    StableDiffusionMixin,
+    TextualInversionLoaderMixin,
+    IPAdapterMixin,
+    StableDiffusionLoraLoaderMixin,
+    FromSingleFileMixin,
+):
+    r"""
+    Pipeline for text-guided image-to-image generation using Alt Diffusion.
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods
+    implemented for all pipelines (downloading, saving, running on a particular device, etc.).
+
+    The pipeline also inherits the following loading methods:
+        - [`~loaders.TextualInversionLoaderMixin.load_textual_inversion`] for loading textual inversion embeddings
+        - [`~loaders.StableDiffusionLoraLoaderMixin.load_lora_weights`] for loading LoRA weights
+        - [`~loaders.StableDiffusionLoraLoaderMixin.save_lora_weights`] for saving LoRA weights
+        - [`~loaders.FromSingleFileMixin.from_single_file`] for loading `.ckpt` files
+        - [`~loaders.IPAdapterMixin.load_ip_adapter`] for loading IP Adapters
+
+    Args:
+        vae ([`AutoencoderKL`]):
+            Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations.
+        text_encoder ([`~transformers.RobertaSeriesModelWithTransformation`]):
+            Frozen text-encoder ([clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14)).
+        tokenizer ([`~transformers.XLMRobertaTokenizer`]):
+            A `XLMRobertaTokenizer` to tokenize text.
+        unet ([`UNet2DConditionModel`]):
+            A `UNet2DConditionModel` to denoise the encoded image latents.
+        scheduler ([`SchedulerMixin`]):
+            A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of
+            [`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`].
+        safety_checker ([`StableDiffusionSafetyChecker`]):
+            Classification module that estimates whether generated images could be considered offensive or harmful.
+            Please refer to the [model card](https://huggingface.co/runwayml/stable-diffusion-v1-5) for more details
+            about a model's potential harms.
+        feature_extractor ([`~transformers.CLIPImageProcessor`]):
+            A `CLIPImageProcessor` to extract features from generated images; used as inputs to the `safety_checker`.
+    """
+
+    model_cpu_offload_seq = "text_encoder->image_encoder->unet->vae"
+    _optional_components = ["safety_checker", "feature_extractor", "image_encoder"]
+    _exclude_from_cpu_offload = ["safety_checker"]
+    _callback_tensor_inputs = ["latents", "prompt_embeds", "negative_prompt_embeds"]
+
+    def __init__(
+        self,
+        vae: AutoencoderKL,
+        text_encoder: RobertaSeriesModelWithTransformation,
+        tokenizer: XLMRobertaTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: KarrasDiffusionSchedulers,
+        safety_checker: StableDiffusionSafetyChecker,
+        feature_extractor: CLIPImageProcessor,
+        image_encoder: CLIPVisionModelWithProjection = None,
+        requires_safety_checker: bool = True,
+    ):
+        super().__init__()
+
+        if scheduler is not None and getattr(scheduler.config, "steps_offset", 1) != 1:
+            deprecation_message = (
+                f"The configuration file of this scheduler: {scheduler} is outdated. `steps_offset`"
+                f" should be set to 1 instead of {scheduler.config.steps_offset}. Please make sure "
+                "to update the config accordingly as leaving `steps_offset` might led to incorrect results"
+                " in future versions. If you have downloaded this checkpoint from the Hugging Face Hub,"
+                " it would be very nice if you could open a Pull request for the `scheduler/scheduler_config.json`"
+                " file"
+            )
+            deprecate("steps_offset!=1", "1.0.0", deprecation_message, standard_warn=False)
+            new_config = dict(scheduler.config)
+            new_config["steps_offset"] = 1
+            scheduler._internal_dict = FrozenDict(new_config)
+
+        if scheduler is not None and getattr(scheduler.config, "clip_sample", False) is True:
+            deprecation_message = (
+                f"The configuration file of this scheduler: {scheduler} has not set the configuration `clip_sample`."
+                " `clip_sample` should be set to False in the configuration file. Please make sure to update the"
+                " config accordingly as not setting `clip_sample` in the config might lead to incorrect results in"
+                " future versions. If you have downloaded this checkpoint from the Hugging Face Hub, it would be very"
+                " nice if you could open a Pull request for the `scheduler/scheduler_config.json` file"
+            )
+            deprecate("clip_sample not set", "1.0.0", deprecation_message, standard_warn=False)
+            new_config = dict(scheduler.config)
+            new_config["clip_sample"] = False
+            scheduler._internal_dict = FrozenDict(new_config)
+
+        if safety_checker is None and requires_safety_checker:
+            logger.warning(
+                f"You have disabled the safety checker for {self.__class__} by passing `safety_checker=None`. Ensure"
+                " that you abide to the conditions of the Alt Diffusion license and do not expose unfiltered"
+                " results in services or applications open to the public. Both the diffusers team and Hugging Face"
+                " strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling"
+                " it only for use-cases that involve analyzing network behavior or auditing its results. For more"
+                " information, please have a look at https://github.com/huggingface/diffusers/pull/254 ."
+            )
+
+        if safety_checker is not None and feature_extractor is None:
+            raise ValueError(
+                "Make sure to define a feature extractor when loading {self.__class__} if you want to use the safety"
+                " checker. If you do not want to use the safety checker, you can pass `'safety_checker=None'` instead."
+            )
+
+        is_unet_version_less_0_9_0 = (
+            unet is not None
+            and hasattr(unet.config, "_diffusers_version")
+            and version.parse(version.parse(unet.config._diffusers_version).base_version) < version.parse("0.9.0.dev0")
+        )
+        is_unet_sample_size_less_64 = (
+            unet is not None and hasattr(unet.config, "sample_size") and unet.config.sample_size < 64
+        )
+        if is_unet_version_less_0_9_0 and is_unet_sample_size_less_64:
+            deprecation_message = (
+                "The configuration file of the unet has set the default `sample_size` to smaller than"
+                " 64 which seems highly unlikely. If your checkpoint is a fine-tuned version of any of the"
+                " following: \n- CompVis/stable-diffusion-v1-4 \n- CompVis/stable-diffusion-v1-3 \n-"
+                " CompVis/stable-diffusion-v1-2 \n- CompVis/stable-diffusion-v1-1 \n- runwayml/stable-diffusion-v1-5"
+                " \n- runwayml/stable-diffusion-inpainting \n you should change 'sample_size' to 64 in the"
+                " configuration file. Please make sure to update the config accordingly as leaving `sample_size=32`"
+                " in the config might lead to incorrect results in future versions. If you have downloaded this"
+                " checkpoint from the Hugging Face Hub, it would be very nice if you could open a Pull request for"
+                " the `unet/config.json` file"
+            )
+            deprecate("sample_size<64", "1.0.0", deprecation_message, standard_warn=False)
+            new_config = dict(unet.config)
+            new_config["sample_size"] = 64
+            unet._internal_dict = FrozenDict(new_config)
+
+        self.register_modules(
+            vae=vae,
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            safety_checker=safety_checker,
+            feature_extractor=feature_extractor,
+            image_encoder=image_encoder,
+        )
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1) if getattr(self, "vae", None) else 8
+        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self.register_to_config(requires_safety_checker=requires_safety_checker)
+
+    def _encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        lora_scale: Optional[float] = None,
+        **kwargs,
+    ):
+        deprecation_message = "`_encode_prompt()` is deprecated and it will be removed in a future version. Use `encode_prompt()` instead. Also, be aware that the output format changed from a concatenated tensor to a tuple."
+        deprecate("_encode_prompt()", "1.0.0", deprecation_message, standard_warn=False)
+
+        prompt_embeds_tuple = self.encode_prompt(
+            prompt=prompt,
+            device=device,
+            num_images_per_prompt=num_images_per_prompt,
+            do_classifier_free_guidance=do_classifier_free_guidance,
+            negative_prompt=negative_prompt,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
+            lora_scale=lora_scale,
+            **kwargs,
+        )
+
+        # concatenate for backwards comp
+        prompt_embeds = torch.cat([prompt_embeds_tuple[1], prompt_embeds_tuple[0]])
+
+        return prompt_embeds
+
+    def encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        lora_scale: Optional[float] = None,
+        clip_skip: Optional[int] = None,
+    ):
+        r"""
+        Encodes the prompt into text encoder hidden states.
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                prompt to be encoded
+            device: (`torch.device`):
+                torch device
+            num_images_per_prompt (`int`):
+                number of images that should be generated per prompt
+            do_classifier_free_guidance (`bool`):
+                whether to use classifier free guidance or not
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. If not defined, one has to pass
+                `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
+                less than `1`).
+            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.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.
+            lora_scale (`float`, *optional*):
+                A LoRA scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded.
+            clip_skip (`int`, *optional*):
+                Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
+                the output of the pre-final layer will be used for computing the prompt embeddings.
+        """
+        # set lora scale so that monkey patched LoRA
+        # function of text encoder can correctly access it
+        if lora_scale is not None and isinstance(self, StableDiffusionLoraLoaderMixin):
+            self._lora_scale = lora_scale
+
+            # dynamically adjust the LoRA scale
+            if not USE_PEFT_BACKEND:
+                adjust_lora_scale_text_encoder(self.text_encoder, lora_scale)
+            else:
+                scale_lora_layers(self.text_encoder, lora_scale)
+
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        if prompt_embeds is None:
+            # textual inversion: process multi-vector tokens if necessary
+            if isinstance(self, TextualInversionLoaderMixin):
+                prompt = self.maybe_convert_prompt(prompt, self.tokenizer)
+
+            text_inputs = self.tokenizer(
+                prompt,
+                padding="max_length",
+                max_length=self.tokenizer.model_max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+            text_input_ids = text_inputs.input_ids
+            untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+            if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(
+                text_input_ids, untruncated_ids
+            ):
+                removed_text = self.tokenizer.batch_decode(
+                    untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1]
+                )
+                logger.warning(
+                    "The following part of your input was truncated because CLIP can only handle sequences up to"
+                    f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+                )
+
+            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+                attention_mask = text_inputs.attention_mask.to(device)
+            else:
+                attention_mask = None
+
+            if clip_skip is None:
+                prompt_embeds = self.text_encoder(text_input_ids.to(device), attention_mask=attention_mask)
+                prompt_embeds = prompt_embeds[0]
+            else:
+                prompt_embeds = self.text_encoder(
+                    text_input_ids.to(device), attention_mask=attention_mask, output_hidden_states=True
+                )
+                # Access the `hidden_states` first, that contains a tuple of
+                # all the hidden states from the encoder layers. Then index into
+                # the tuple to access the hidden states from the desired layer.
+                prompt_embeds = prompt_embeds[-1][-(clip_skip + 1)]
+                # We also need to apply the final LayerNorm here to not mess with the
+                # representations. The `last_hidden_states` that we typically use for
+                # obtaining the final prompt representations passes through the LayerNorm
+                # layer.
+                prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)
+
+        if self.text_encoder is not None:
+            prompt_embeds_dtype = self.text_encoder.dtype
+        elif self.unet is not None:
+            prompt_embeds_dtype = self.unet.dtype
+        else:
+            prompt_embeds_dtype = prompt_embeds.dtype
+
+        prompt_embeds = prompt_embeds.to(dtype=prompt_embeds_dtype, device=device)
+
+        bs_embed, seq_len, _ = prompt_embeds.shape
+        # duplicate text embeddings for each generation per prompt, using mps friendly method
+        prompt_embeds = prompt_embeds.repeat(1, num_images_per_prompt, 1)
+        prompt_embeds = prompt_embeds.view(bs_embed * num_images_per_prompt, seq_len, -1)
+
+        # get unconditional embeddings for classifier free guidance
+        if do_classifier_free_guidance and negative_prompt_embeds is None:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif prompt is not None and type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            # textual inversion: process multi-vector tokens if necessary
+            if isinstance(self, TextualInversionLoaderMixin):
+                uncond_tokens = self.maybe_convert_prompt(uncond_tokens, self.tokenizer)
+
+            max_length = prompt_embeds.shape[1]
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+
+            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+                attention_mask = uncond_input.attention_mask.to(device)
+            else:
+                attention_mask = None
+
+            negative_prompt_embeds = self.text_encoder(
+                uncond_input.input_ids.to(device),
+                attention_mask=attention_mask,
+            )
+            negative_prompt_embeds = negative_prompt_embeds[0]
+
+        if do_classifier_free_guidance:
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+            seq_len = negative_prompt_embeds.shape[1]
+
+            negative_prompt_embeds = negative_prompt_embeds.to(dtype=prompt_embeds_dtype, device=device)
+
+            negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt, 1)
+            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)
+
+        if isinstance(self, StableDiffusionLoraLoaderMixin) 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
+
+    def encode_image(self, image, device, num_images_per_prompt, output_hidden_states=None):
+        dtype = next(self.image_encoder.parameters()).dtype
+
+        if not isinstance(image, torch.Tensor):
+            image = self.feature_extractor(image, return_tensors="pt").pixel_values
+
+        image = image.to(device=device, dtype=dtype)
+        if output_hidden_states:
+            image_enc_hidden_states = self.image_encoder(image, output_hidden_states=True).hidden_states[-2]
+            image_enc_hidden_states = image_enc_hidden_states.repeat_interleave(num_images_per_prompt, dim=0)
+            uncond_image_enc_hidden_states = self.image_encoder(
+                torch.zeros_like(image), output_hidden_states=True
+            ).hidden_states[-2]
+            uncond_image_enc_hidden_states = uncond_image_enc_hidden_states.repeat_interleave(
+                num_images_per_prompt, dim=0
+            )
+            return image_enc_hidden_states, uncond_image_enc_hidden_states
+        else:
+            image_embeds = self.image_encoder(image).image_embeds
+            image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+            uncond_image_embeds = torch.zeros_like(image_embeds)
+
+            return image_embeds, uncond_image_embeds
+
+    def run_safety_checker(self, image, device, dtype):
+        if self.safety_checker is None:
+            has_nsfw_concept = None
+        else:
+            if torch.is_tensor(image):
+                feature_extractor_input = self.image_processor.postprocess(image, output_type="pil")
+            else:
+                feature_extractor_input = self.image_processor.numpy_to_pil(image)
+            safety_checker_input = self.feature_extractor(feature_extractor_input, return_tensors="pt").to(device)
+            image, has_nsfw_concept = self.safety_checker(
+                images=image, clip_input=safety_checker_input.pixel_values.to(dtype)
+            )
+        return image, has_nsfw_concept
+
+    def decode_latents(self, latents):
+        deprecation_message = "The decode_latents method is deprecated and will be removed in 1.0.0. Please use VaeImageProcessor.postprocess(...) instead"
+        deprecate("decode_latents", "1.0.0", deprecation_message, standard_warn=False)
+
+        latents = 1 / self.vae.config.scaling_factor * latents
+        image = self.vae.decode(latents, return_dict=False)[0]
+        image = (image / 2 + 0.5).clamp(0, 1)
+        # we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
+        image = image.cpu().permute(0, 2, 3, 1).float().numpy()
+        return image
+
+    def prepare_extra_step_kwargs(self, generator, eta):
+        # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
+        # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
+        # eta corresponds to η in DDIM paper: https://huggingface.co/papers/2010.02502
+        # and should be between [0, 1]
+
+        accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        extra_step_kwargs = {}
+        if accepts_eta:
+            extra_step_kwargs["eta"] = eta
+
+        # check if the scheduler accepts generator
+        accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        if accepts_generator:
+            extra_step_kwargs["generator"] = generator
+        return extra_step_kwargs
+
+    def check_inputs(
+        self,
+        prompt,
+        strength,
+        callback_steps,
+        negative_prompt=None,
+        prompt_embeds=None,
+        negative_prompt_embeds=None,
+        callback_on_step_end_tensor_inputs=None,
+    ):
+        if strength < 0 or strength > 1:
+            raise ValueError(f"The value of strength should in [0.0, 1.0] but is {strength}")
+
+        if callback_steps is not None and (not isinstance(callback_steps, int) or callback_steps <= 0):
+            raise ValueError(
+                f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
+                f" {type(callback_steps)}."
+            )
+
+        if callback_on_step_end_tensor_inputs is not None and not all(
+            k in self._callback_tensor_inputs for k in callback_on_step_end_tensor_inputs
+        ):
+            raise ValueError(
+                f"`callback_on_step_end_tensor_inputs` has to be in {self._callback_tensor_inputs}, but found {[k for k in callback_on_step_end_tensor_inputs if k not in self._callback_tensor_inputs]}"
+            )
+        if prompt is not None and prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
+                " only forward one of the two."
+            )
+        elif prompt is None and prompt_embeds is None:
+            raise ValueError(
+                "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
+            )
+        elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        if negative_prompt is not None and negative_prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
+                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
+            )
+
+        if prompt_embeds is not None and negative_prompt_embeds is not None:
+            if prompt_embeds.shape != negative_prompt_embeds.shape:
+                raise ValueError(
+                    "`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
+                    f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
+                    f" {negative_prompt_embeds.shape}."
+                )
+
+    def get_timesteps(self, num_inference_steps, strength, device):
+        # get the original timestep using init_timestep
+        init_timestep = min(int(num_inference_steps * strength), num_inference_steps)
+
+        t_start = max(num_inference_steps - init_timestep, 0)
+        timesteps = self.scheduler.timesteps[t_start * self.scheduler.order :]
+
+        return timesteps, num_inference_steps - t_start
+
+    def prepare_latents(self, image, timestep, batch_size, num_images_per_prompt, dtype, device, generator=None):
+        if not isinstance(image, (torch.Tensor, PIL.Image.Image, list)):
+            raise ValueError(
+                f"`image` has to be of type `torch.Tensor`, `PIL.Image.Image` or list but is {type(image)}"
+            )
+
+        image = image.to(device=device, dtype=dtype)
+
+        batch_size = batch_size * num_images_per_prompt
+
+        if image.shape[1] == 4:
+            init_latents = image
+
+        else:
+            if isinstance(generator, list) and len(generator) != batch_size:
+                raise ValueError(
+                    f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
+                    f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+                )
+
+            elif isinstance(generator, list):
+                init_latents = [
+                    retrieve_latents(self.vae.encode(image[i : i + 1]), generator=generator[i])
+                    for i in range(batch_size)
+                ]
+                init_latents = torch.cat(init_latents, dim=0)
+            else:
+                init_latents = retrieve_latents(self.vae.encode(image), generator=generator)
+
+            init_latents = self.vae.config.scaling_factor * init_latents
+
+        if batch_size > init_latents.shape[0] and batch_size % init_latents.shape[0] == 0:
+            # expand init_latents for batch_size
+            deprecation_message = (
+                f"You have passed {batch_size} text prompts (`prompt`), but only {init_latents.shape[0]} initial"
+                " images (`image`). Initial images are now duplicating to match the number of text prompts. Note"
+                " that this behavior is deprecated and will be removed in a version 1.0.0. Please make sure to update"
+                " your script to pass as many initial images as text prompts to suppress this warning."
+            )
+            deprecate("len(prompt) != len(image)", "1.0.0", deprecation_message, standard_warn=False)
+            additional_image_per_prompt = batch_size // init_latents.shape[0]
+            init_latents = torch.cat([init_latents] * additional_image_per_prompt, dim=0)
+        elif batch_size > init_latents.shape[0] and batch_size % init_latents.shape[0] != 0:
+            raise ValueError(
+                f"Cannot duplicate `image` of batch size {init_latents.shape[0]} to {batch_size} text prompts."
+            )
+        else:
+            init_latents = torch.cat([init_latents], dim=0)
+
+        shape = init_latents.shape
+        noise = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+
+        # get latents
+        init_latents = self.scheduler.add_noise(init_latents, noise, timestep)
+        latents = init_latents
+
+        return latents
+
+    def get_guidance_scale_embedding(self, w, embedding_dim=512, dtype=torch.float32):
+        """
+        See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
+
+        Args:
+            timesteps (`torch.Tensor`):
+                generate embedding vectors at these timesteps
+            embedding_dim (`int`, *optional*, defaults to 512):
+                dimension of the embeddings to generate
+            dtype:
+                data type of the generated embeddings
+
+        Returns:
+            `torch.Tensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
+        """
+        assert len(w.shape) == 1
+        w = w * 1000.0
+
+        half_dim = embedding_dim // 2
+        emb = torch.log(torch.tensor(10000.0)) / (half_dim - 1)
+        emb = torch.exp(torch.arange(half_dim, dtype=dtype) * -emb)
+        emb = w.to(dtype)[:, None] * emb[None, :]
+        emb = torch.cat([torch.sin(emb), torch.cos(emb)], dim=1)
+        if embedding_dim % 2 == 1:  # zero pad
+            emb = torch.nn.functional.pad(emb, (0, 1))
+        assert emb.shape == (w.shape[0], embedding_dim)
+        return emb
+
+    @property
+    def guidance_scale(self):
+        return self._guidance_scale
+
+    @property
+    def clip_skip(self):
+        return self._clip_skip
+
+    # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
+    # of the Imagen paper: https://huggingface.co/papers/2205.11487 . `guidance_scale = 1`
+    # corresponds to doing no classifier free guidance.
+    @property
+    def do_classifier_free_guidance(self):
+        return self._guidance_scale > 1 and self.unet.config.time_cond_proj_dim is None
+
+    @property
+    def cross_attention_kwargs(self):
+        return self._cross_attention_kwargs
+
+    @property
+    def num_timesteps(self):
+        return self._num_timesteps
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]] = None,
+        image: PipelineImageInput = None,
+        strength: float = 0.8,
+        num_inference_steps: Optional[int] = 50,
+        timesteps: List[int] = None,
+        sigmas: List[float] = None,
+        guidance_scale: Optional[float] = 7.5,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        num_images_per_prompt: Optional[int] = 1,
+        eta: Optional[float] = 0.0,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        ip_adapter_image: Optional[PipelineImageInput] = None,
+        output_type: Optional[str] = "pil",
+        return_dict: bool = True,
+        cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+        clip_skip: int = None,
+        callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
+        callback_on_step_end_tensor_inputs: List[str] = ["latents"],
+        **kwargs,
+    ):
+        r"""
+        The call function to the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
+            image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`):
+                `Image`, numpy array or tensor representing an image batch to be used as the starting point. For both
+                numpy array and pytorch tensor, the expected value range is between `[0, 1]` If it's a tensor or a list
+                or tensors, the expected shape should be `(B, C, H, W)` or `(C, H, W)`. If it is a numpy array or a
+                list of arrays, the expected shape should be `(B, H, W, C)` or `(H, W, C)` It can also accept image
+                latents as `image`, but if passing latents directly it is not encoded again.
+            strength (`float`, *optional*, defaults to 0.8):
+                Indicates extent to transform the reference `image`. Must be between 0 and 1. `image` is used as a
+                starting point and more noise is added the higher the `strength`. The number of denoising steps depends
+                on the amount of noise initially added. When `strength` is 1, added noise is maximum and the denoising
+                process runs for the full number of iterations specified in `num_inference_steps`. A value of 1
+                essentially ignores `image`.
+            num_inference_steps (`int`, *optional*, defaults to 50):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference. This parameter is modulated by `strength`.
+            timesteps (`List[int]`, *optional*):
+                Custom timesteps to use for the denoising process with schedulers which support a `timesteps` argument
+                in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is
+                passed will be used. Must be in descending order.
+            guidance_scale (`float`, *optional*, defaults to 7.5):
+                A higher guidance scale value encourages the model to generate images closely linked to the text
+                `prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide what to not include in image generation. If not defined, you need to
+                pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`).
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            eta (`float`, *optional*, defaults to 0.0):
+                Corresponds to parameter eta (η) from the [DDIM](https://huggingface.co/papers/2010.02502) paper. Only
+                applies to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers.
+            generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
+                A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
+                generation deterministic.
+            prompt_embeds (`torch.Tensor`, *optional*):
+                Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
+                provided, text embeddings are generated from the `prompt` input argument.
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
+                Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
+                not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
+            ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generated image. Choose between `PIL.Image` or `np.array`.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.stable_diffusion.AltDiffusionPipelineOutput`] instead of a
+                plain tuple.
+            cross_attention_kwargs (`dict`, *optional*):
+                A kwargs dictionary that if specified is passed along to the [`AttentionProcessor`] as defined in
+                [`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+            clip_skip (`int`, *optional*):
+                Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
+                the output of the pre-final layer will be used for computing the prompt embeddings.
+            callback_on_step_end (`Callable`, *optional*):
+                A function that calls at the end of each denoising steps during the 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_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
+                `._callback_tensor_inputs` attribute of your pipeline class.
+        Examples:
+
+        Returns:
+            [`~pipelines.stable_diffusion.AltDiffusionPipelineOutput`] or `tuple`:
+                If `return_dict` is `True`, [`~pipelines.stable_diffusion.AltDiffusionPipelineOutput`] is returned,
+                otherwise a `tuple` is returned where the first element is a list with the generated images and the
+                second element is a list of `bool`s indicating whether the corresponding generated image contains
+                "not-safe-for-work" (nsfw) content.
+        """
+
+        callback = kwargs.pop("callback", None)
+        callback_steps = kwargs.pop("callback_steps", None)
+
+        if callback is not None:
+            deprecate(
+                "callback",
+                "1.0.0",
+                "Passing `callback` as an input argument to `__call__` is deprecated, consider use `callback_on_step_end`",
+            )
+        if callback_steps is not None:
+            deprecate(
+                "callback_steps",
+                "1.0.0",
+                "Passing `callback_steps` as an input argument to `__call__` is deprecated, consider use `callback_on_step_end`",
+            )
+
+        # 1. Check inputs. Raise error if not correct
+        self.check_inputs(
+            prompt,
+            strength,
+            callback_steps,
+            negative_prompt,
+            prompt_embeds,
+            negative_prompt_embeds,
+            callback_on_step_end_tensor_inputs,
+        )
+
+        self._guidance_scale = guidance_scale
+        self._clip_skip = clip_skip
+        self._cross_attention_kwargs = cross_attention_kwargs
+
+        # 2. Define call parameters
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        device = self._execution_device
+
+        # 3. Encode input prompt
+        text_encoder_lora_scale = (
+            self.cross_attention_kwargs.get("scale", None) if self.cross_attention_kwargs is not None else None
+        )
+        prompt_embeds, negative_prompt_embeds = self.encode_prompt(
+            prompt,
+            device,
+            num_images_per_prompt,
+            self.do_classifier_free_guidance,
+            negative_prompt,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
+            lora_scale=text_encoder_lora_scale,
+            clip_skip=self.clip_skip,
+        )
+        # For classifier free guidance, we need to do two forward passes.
+        # Here we concatenate the unconditional and text embeddings into a single batch
+        # to avoid doing two forward passes
+        if self.do_classifier_free_guidance:
+            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
+
+        if ip_adapter_image is not None:
+            output_hidden_state = False if isinstance(self.unet.encoder_hid_proj, ImageProjection) else True
+            image_embeds, negative_image_embeds = self.encode_image(
+                ip_adapter_image, device, num_images_per_prompt, output_hidden_state
+            )
+            if self.do_classifier_free_guidance:
+                image_embeds = torch.cat([negative_image_embeds, image_embeds])
+
+        # 4. Preprocess image
+        image = self.image_processor.preprocess(image)
+
+        # 5. set timesteps
+        timesteps, num_inference_steps = retrieve_timesteps(
+            self.scheduler, num_inference_steps, device, timesteps, sigmas
+        )
+        timesteps, num_inference_steps = self.get_timesteps(num_inference_steps, strength, device)
+        latent_timestep = timesteps[:1].repeat(batch_size * num_images_per_prompt)
+
+        # 6. Prepare latent variables
+        latents = self.prepare_latents(
+            image,
+            latent_timestep,
+            batch_size,
+            num_images_per_prompt,
+            prompt_embeds.dtype,
+            device,
+            generator,
+        )
+
+        # 7. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
+        extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)
+
+        # 7.1 Add image embeds for IP-Adapter
+        added_cond_kwargs = {"image_embeds": image_embeds} if ip_adapter_image is not None else None
+
+        # 7.2 Optionally get Guidance Scale Embedding
+        timestep_cond = None
+        if self.unet.config.time_cond_proj_dim is not None:
+            guidance_scale_tensor = torch.tensor(self.guidance_scale - 1).repeat(batch_size * num_images_per_prompt)
+            timestep_cond = self.get_guidance_scale_embedding(
+                guidance_scale_tensor, embedding_dim=self.unet.config.time_cond_proj_dim
+            ).to(device=device, dtype=latents.dtype)
+
+        # 8. Denoising loop
+        num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order
+        self._num_timesteps = len(timesteps)
+        with self.progress_bar(total=num_inference_steps) as progress_bar:
+            for i, t in enumerate(timesteps):
+                # expand the latents if we are doing classifier free guidance
+                latent_model_input = torch.cat([latents] * 2) if self.do_classifier_free_guidance else latents
+                latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
+
+                # predict the noise residual
+                noise_pred = self.unet(
+                    latent_model_input,
+                    t,
+                    encoder_hidden_states=prompt_embeds,
+                    timestep_cond=timestep_cond,
+                    cross_attention_kwargs=self.cross_attention_kwargs,
+                    added_cond_kwargs=added_cond_kwargs,
+                    return_dict=False,
+                )[0]
+
+                # perform guidance
+                if self.do_classifier_free_guidance:
+                    noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
+                    noise_pred = noise_pred_uncond + self.guidance_scale * (noise_pred_text - noise_pred_uncond)
+
+                # compute the previous noisy sample x_t -> x_t-1
+                latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs, return_dict=False)[0]
+
+                if callback_on_step_end is not None:
+                    callback_kwargs = {}
+                    for k in callback_on_step_end_tensor_inputs:
+                        callback_kwargs[k] = locals()[k]
+                    callback_outputs = callback_on_step_end(self, i, t, callback_kwargs)
+
+                    latents = callback_outputs.pop("latents", latents)
+                    prompt_embeds = callback_outputs.pop("prompt_embeds", prompt_embeds)
+                    negative_prompt_embeds = callback_outputs.pop("negative_prompt_embeds", negative_prompt_embeds)
+
+                # call the callback, if provided
+                if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
+                    progress_bar.update()
+                    if callback is not None and i % callback_steps == 0:
+                        step_idx = i // getattr(self.scheduler, "order", 1)
+                        callback(step_idx, t, latents)
+
+        if not output_type == "latent":
+            image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False, generator=generator)[
+                0
+            ]
+            image, has_nsfw_concept = self.run_safety_checker(image, device, prompt_embeds.dtype)
+        else:
+            image = latents
+            has_nsfw_concept = None
+
+        if has_nsfw_concept is None:
+            do_denormalize = [True] * image.shape[0]
+        else:
+            do_denormalize = [not has_nsfw for has_nsfw in has_nsfw_concept]
+
+        image = self.image_processor.postprocess(image, output_type=output_type, do_denormalize=do_denormalize)
+
+        # Offload all models
+        self.maybe_free_model_hooks()
+
+        if not return_dict:
+            return (image, has_nsfw_concept)
+
+        return AltDiffusionPipelineOutput(images=image, nsfw_content_detected=has_nsfw_concept)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/deprecated/alt_diffusion/pipeline_output.py

@@ -0,0 +1,28 @@
+from dataclasses import dataclass
+from typing import List, Optional, Union
+
+import numpy as np
+import PIL.Image
+
+from ....utils import (
+    BaseOutput,
+)
+
+
+@dataclass
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_output.StableDiffusionPipelineOutput with Stable->Alt
+class AltDiffusionPipelineOutput(BaseOutput):
+    """
+    Output class for Alt Diffusion pipelines.
+
+    Args:
+        images (`List[PIL.Image.Image]` or `np.ndarray`)
+            List of denoised PIL images of length `batch_size` or NumPy array of shape `(batch_size, height, width,
+            num_channels)`.
+        nsfw_content_detected (`List[bool]`)
+            List indicating whether the corresponding generated image contains "not-safe-for-work" (nsfw) content or
+            `None` if safety checking could not be performed.
+    """
+
+    images: Union[List[PIL.Image.Image], np.ndarray]
+    nsfw_content_detected: Optional[List[bool]]

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/hunyuan_video/__init__.py

@@ -0,0 +1,54 @@
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    DIFFUSERS_SLOW_IMPORT,
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    get_objects_from_module,
+    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_hunyuan_skyreels_image2video"] = ["HunyuanSkyreelsImageToVideoPipeline"]
+    _import_structure["pipeline_hunyuan_video"] = ["HunyuanVideoPipeline"]
+    _import_structure["pipeline_hunyuan_video_framepack"] = ["HunyuanVideoFramepackPipeline"]
+    _import_structure["pipeline_hunyuan_video_image2video"] = ["HunyuanVideoImageToVideoPipeline"]
+
+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_hunyuan_skyreels_image2video import HunyuanSkyreelsImageToVideoPipeline
+        from .pipeline_hunyuan_video import HunyuanVideoPipeline
+        from .pipeline_hunyuan_video_framepack import HunyuanVideoFramepackPipeline
+        from .pipeline_hunyuan_video_image2video import HunyuanVideoImageToVideoPipeline
+
+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)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/hunyuan_video/pipeline_output.py

@@ -0,0 +1,39 @@
+from dataclasses import dataclass
+from typing import List, Union
+
+import numpy as np
+import PIL.Image
+import torch
+
+from diffusers.utils import BaseOutput
+
+
+@dataclass
+class HunyuanVideoPipelineOutput(BaseOutput):
+    r"""
+    Output class for HunyuanVideo pipelines.
+
+    Args:
+        frames (`torch.Tensor`, `np.ndarray`, or List[List[PIL.Image.Image]]):
+            List of video outputs - It can be a nested list of length `batch_size,` with each sub-list containing
+            denoised PIL image sequences of length `num_frames.` It can also be a NumPy array or Torch tensor of shape
+            `(batch_size, num_frames, channels, height, width)`.
+    """
+
+    frames: torch.Tensor
+
+
+@dataclass
+class HunyuanVideoFramepackPipelineOutput(BaseOutput):
+    r"""
+    Output class for HunyuanVideo pipelines.
+
+    Args:
+        frames (`torch.Tensor`, `np.ndarray`, or List[List[PIL.Image.Image]]):
+            List of video outputs - It can be a nested list of length `batch_size,` with each sub-list containing
+            denoised PIL image sequences of length `num_frames.` It can also be a NumPy array or Torch tensor of shape
+            `(batch_size, num_frames, channels, height, width)`. Or, a list of torch tensors where each tensor
+            corresponds to a latent that decodes to multiple frames.
+    """
+
+    frames: Union[torch.Tensor, np.ndarray, List[List[PIL.Image.Image]], List[torch.Tensor]]

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/i2vgen_xl/__init__.py

@@ -0,0 +1,46 @@
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    DIFFUSERS_SLOW_IMPORT,
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    get_objects_from_module,
+    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_i2vgen_xl"] = ["I2VGenXLPipeline"]
+
+
+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 *  # noqa F403
+    else:
+        from .pipeline_i2vgen_xl import I2VGenXLPipeline
+
+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)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/i2vgen_xl/pipeline_i2vgen_xl.py

@@ -0,0 +1,797 @@
+# Copyright 2025 Alibaba DAMO-VILAB and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import inspect
+from dataclasses import dataclass
+from typing import Any, Dict, List, Optional, Tuple, Union
+
+import numpy as np
+import PIL
+import torch
+from transformers import CLIPImageProcessor, CLIPTextModel, CLIPTokenizer, CLIPVisionModelWithProjection
+
+from ...image_processor import PipelineImageInput, VaeImageProcessor
+from ...models import AutoencoderKL
+from ...models.unets.unet_i2vgen_xl import I2VGenXLUNet
+from ...schedulers import DDIMScheduler
+from ...utils import (
+    BaseOutput,
+    is_torch_xla_available,
+    logging,
+    replace_example_docstring,
+)
+from ...utils.torch_utils import randn_tensor
+from ...video_processor import VideoProcessor
+from ..pipeline_utils import DeprecatedPipelineMixin, DiffusionPipeline, StableDiffusionMixin
+
+
+if is_torch_xla_available():
+    import torch_xla.core.xla_model as xm
+
+    XLA_AVAILABLE = True
+else:
+    XLA_AVAILABLE = False
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+
+EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> import torch
+        >>> from diffusers import I2VGenXLPipeline
+        >>> from diffusers.utils import export_to_gif, load_image
+
+        >>> pipeline = I2VGenXLPipeline.from_pretrained(
+        ...     "ali-vilab/i2vgen-xl", torch_dtype=torch.float16, variant="fp16"
+        ... )
+        >>> pipeline.enable_model_cpu_offload()
+
+        >>> image_url = (
+        ...     "https://huggingface.co/datasets/diffusers/docs-images/resolve/main/i2vgen_xl_images/img_0009.png"
+        ... )
+        >>> image = load_image(image_url).convert("RGB")
+
+        >>> prompt = "Papers were floating in the air on a table in the library"
+        >>> negative_prompt = "Distorted, discontinuous, Ugly, blurry, low resolution, motionless, static, disfigured, disconnected limbs, Ugly faces, incomplete arms"
+        >>> generator = torch.manual_seed(8888)
+
+        >>> frames = pipeline(
+        ...     prompt=prompt,
+        ...     image=image,
+        ...     num_inference_steps=50,
+        ...     negative_prompt=negative_prompt,
+        ...     guidance_scale=9.0,
+        ...     generator=generator,
+        ... ).frames[0]
+        >>> video_path = export_to_gif(frames, "i2v.gif")
+        ```
+"""
+
+
+@dataclass
+class I2VGenXLPipelineOutput(BaseOutput):
+    r"""
+     Output class for image-to-video pipeline.
+
+    Args:
+         frames (`torch.Tensor`, `np.ndarray`, or List[List[PIL.Image.Image]]):
+             List of video outputs - It can be a nested list of length `batch_size,` with each sub-list containing
+             denoised
+     PIL image sequences of length `num_frames.` It can also be a NumPy array or Torch tensor of shape
+    `(batch_size, num_frames, channels, height, width)`
+    """
+
+    frames: Union[torch.Tensor, np.ndarray, List[List[PIL.Image.Image]]]
+
+
+class I2VGenXLPipeline(
+    DeprecatedPipelineMixin,
+    DiffusionPipeline,
+    StableDiffusionMixin,
+):
+    _last_supported_version = "0.33.1"
+    r"""
+    Pipeline for image-to-video generation as proposed in [I2VGenXL](https://i2vgen-xl.github.io/).
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods
+    implemented for all pipelines (downloading, saving, running on a particular device, etc.).
+
+    Args:
+        vae ([`AutoencoderKL`]):
+            Variational Auto-Encoder (VAE) Model to encode and decode images to and from latent representations.
+        text_encoder ([`CLIPTextModel`]):
+            Frozen text-encoder ([clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14)).
+        tokenizer (`CLIPTokenizer`):
+            A [`~transformers.CLIPTokenizer`] to tokenize text.
+        unet ([`I2VGenXLUNet`]):
+            A [`I2VGenXLUNet`] to denoise the encoded video latents.
+        scheduler ([`DDIMScheduler`]):
+            A scheduler to be used in combination with `unet` to denoise the encoded image latents.
+    """
+
+    model_cpu_offload_seq = "text_encoder->image_encoder->unet->vae"
+
+    def __init__(
+        self,
+        vae: AutoencoderKL,
+        text_encoder: CLIPTextModel,
+        tokenizer: CLIPTokenizer,
+        image_encoder: CLIPVisionModelWithProjection,
+        feature_extractor: CLIPImageProcessor,
+        unet: I2VGenXLUNet,
+        scheduler: DDIMScheduler,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            vae=vae,
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            image_encoder=image_encoder,
+            feature_extractor=feature_extractor,
+            unet=unet,
+            scheduler=scheduler,
+        )
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1) if getattr(self, "vae", None) else 8
+        # `do_resize=False` as we do custom resizing.
+        self.video_processor = VideoProcessor(vae_scale_factor=self.vae_scale_factor, do_resize=False)
+
+    @property
+    def guidance_scale(self):
+        return self._guidance_scale
+
+    # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
+    # of the Imagen paper: https://huggingface.co/papers/2205.11487 . `guidance_scale = 1`
+    # corresponds to doing no classifier free guidance.
+    @property
+    def do_classifier_free_guidance(self):
+        return self._guidance_scale > 1
+
+    def encode_prompt(
+        self,
+        prompt,
+        device,
+        num_videos_per_prompt,
+        negative_prompt=None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        clip_skip: Optional[int] = None,
+    ):
+        r"""
+        Encodes the prompt into text encoder hidden states.
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                prompt to be encoded
+            device: (`torch.device`):
+                torch device
+            num_videos_per_prompt (`int`):
+                number of images that should be generated per prompt
+            do_classifier_free_guidance (`bool`):
+                whether to use classifier free guidance or not
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. If not defined, one has to pass
+                `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
+                less than `1`).
+            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.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.
+            clip_skip (`int`, *optional*):
+                Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
+                the output of the pre-final layer will be used for computing the prompt embeddings.
+        """
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        if prompt_embeds is None:
+            text_inputs = self.tokenizer(
+                prompt,
+                padding="max_length",
+                max_length=self.tokenizer.model_max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+            text_input_ids = text_inputs.input_ids
+            untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+            if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(
+                text_input_ids, untruncated_ids
+            ):
+                removed_text = self.tokenizer.batch_decode(
+                    untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1]
+                )
+                logger.warning(
+                    "The following part of your input was truncated because CLIP can only handle sequences up to"
+                    f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+                )
+
+            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+                attention_mask = text_inputs.attention_mask.to(device)
+            else:
+                attention_mask = None
+
+            if clip_skip is None:
+                prompt_embeds = self.text_encoder(text_input_ids.to(device), attention_mask=attention_mask)
+                prompt_embeds = prompt_embeds[0]
+            else:
+                prompt_embeds = self.text_encoder(
+                    text_input_ids.to(device), attention_mask=attention_mask, output_hidden_states=True
+                )
+                # Access the `hidden_states` first, that contains a tuple of
+                # all the hidden states from the encoder layers. Then index into
+                # the tuple to access the hidden states from the desired layer.
+                prompt_embeds = prompt_embeds[-1][-(clip_skip + 1)]
+                # We also need to apply the final LayerNorm here to not mess with the
+                # representations. The `last_hidden_states` that we typically use for
+                # obtaining the final prompt representations passes through the LayerNorm
+                # layer.
+                prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)
+
+        if self.text_encoder is not None:
+            prompt_embeds_dtype = self.text_encoder.dtype
+        elif self.unet is not None:
+            prompt_embeds_dtype = self.unet.dtype
+        else:
+            prompt_embeds_dtype = prompt_embeds.dtype
+
+        prompt_embeds = prompt_embeds.to(dtype=prompt_embeds_dtype, device=device)
+
+        bs_embed, seq_len, _ = prompt_embeds.shape
+        # duplicate text embeddings for each generation per prompt, using mps friendly method
+        prompt_embeds = prompt_embeds.repeat(1, num_videos_per_prompt, 1)
+        prompt_embeds = prompt_embeds.view(bs_embed * num_videos_per_prompt, seq_len, -1)
+
+        # get unconditional embeddings for classifier free guidance
+        if self.do_classifier_free_guidance and negative_prompt_embeds is None:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif prompt is not None and type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            max_length = prompt_embeds.shape[1]
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+
+            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+                attention_mask = uncond_input.attention_mask.to(device)
+            else:
+                attention_mask = None
+
+            # Apply clip_skip to negative prompt embeds
+            if clip_skip is None:
+                negative_prompt_embeds = self.text_encoder(
+                    uncond_input.input_ids.to(device),
+                    attention_mask=attention_mask,
+                )
+                negative_prompt_embeds = negative_prompt_embeds[0]
+            else:
+                negative_prompt_embeds = self.text_encoder(
+                    uncond_input.input_ids.to(device), attention_mask=attention_mask, output_hidden_states=True
+                )
+                # Access the `hidden_states` first, that contains a tuple of
+                # all the hidden states from the encoder layers. Then index into
+                # the tuple to access the hidden states from the desired layer.
+                negative_prompt_embeds = negative_prompt_embeds[-1][-(clip_skip + 1)]
+                # We also need to apply the final LayerNorm here to not mess with the
+                # representations. The `last_hidden_states` that we typically use for
+                # obtaining the final prompt representations passes through the LayerNorm
+                # layer.
+                negative_prompt_embeds = self.text_encoder.text_model.final_layer_norm(negative_prompt_embeds)
+
+        if self.do_classifier_free_guidance:
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+            seq_len = negative_prompt_embeds.shape[1]
+
+            negative_prompt_embeds = negative_prompt_embeds.to(dtype=prompt_embeds_dtype, device=device)
+
+            negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_videos_per_prompt, 1)
+            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_videos_per_prompt, seq_len, -1)
+
+        return prompt_embeds, negative_prompt_embeds
+
+    def _encode_image(self, image, device, num_videos_per_prompt):
+        dtype = next(self.image_encoder.parameters()).dtype
+
+        if not isinstance(image, torch.Tensor):
+            image = self.video_processor.pil_to_numpy(image)
+            image = self.video_processor.numpy_to_pt(image)
+
+            # Normalize the image with CLIP training stats.
+            image = self.feature_extractor(
+                images=image,
+                do_normalize=True,
+                do_center_crop=False,
+                do_resize=False,
+                do_rescale=False,
+                return_tensors="pt",
+            ).pixel_values
+
+        image = image.to(device=device, dtype=dtype)
+        image_embeddings = self.image_encoder(image).image_embeds
+        image_embeddings = image_embeddings.unsqueeze(1)
+
+        # duplicate image embeddings for each generation per prompt, using mps friendly method
+        bs_embed, seq_len, _ = image_embeddings.shape
+        image_embeddings = image_embeddings.repeat(1, num_videos_per_prompt, 1)
+        image_embeddings = image_embeddings.view(bs_embed * num_videos_per_prompt, seq_len, -1)
+
+        if self.do_classifier_free_guidance:
+            negative_image_embeddings = torch.zeros_like(image_embeddings)
+            image_embeddings = torch.cat([negative_image_embeddings, image_embeddings])
+
+        return image_embeddings
+
+    def decode_latents(self, latents, decode_chunk_size=None):
+        latents = 1 / self.vae.config.scaling_factor * latents
+
+        batch_size, channels, num_frames, height, width = latents.shape
+        latents = latents.permute(0, 2, 1, 3, 4).reshape(batch_size * num_frames, channels, height, width)
+
+        if decode_chunk_size is not None:
+            frames = []
+            for i in range(0, latents.shape[0], decode_chunk_size):
+                frame = self.vae.decode(latents[i : i + decode_chunk_size]).sample
+                frames.append(frame)
+            image = torch.cat(frames, dim=0)
+        else:
+            image = self.vae.decode(latents).sample
+
+        decode_shape = (batch_size, num_frames, -1) + image.shape[2:]
+        video = image[None, :].reshape(decode_shape).permute(0, 2, 1, 3, 4)
+
+        # we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
+        video = video.float()
+        return video
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_extra_step_kwargs
+    def prepare_extra_step_kwargs(self, generator, eta):
+        # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
+        # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
+        # eta corresponds to η in DDIM paper: https://huggingface.co/papers/2010.02502
+        # and should be between [0, 1]
+
+        accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        extra_step_kwargs = {}
+        if accepts_eta:
+            extra_step_kwargs["eta"] = eta
+
+        # check if the scheduler accepts generator
+        accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        if accepts_generator:
+            extra_step_kwargs["generator"] = generator
+        return extra_step_kwargs
+
+    def check_inputs(
+        self,
+        prompt,
+        image,
+        height,
+        width,
+        negative_prompt=None,
+        prompt_embeds=None,
+        negative_prompt_embeds=None,
+    ):
+        if height % 8 != 0 or width % 8 != 0:
+            raise ValueError(f"`height` and `width` have to be divisible by 8 but are {height} and {width}.")
+
+        if prompt is not None and prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
+                " only forward one of the two."
+            )
+        elif prompt is None and prompt_embeds is None:
+            raise ValueError(
+                "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
+            )
+        elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        if negative_prompt is not None and negative_prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
+                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
+            )
+
+        if prompt_embeds is not None and negative_prompt_embeds is not None:
+            if prompt_embeds.shape != negative_prompt_embeds.shape:
+                raise ValueError(
+                    "`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
+                    f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
+                    f" {negative_prompt_embeds.shape}."
+                )
+
+        if (
+            not isinstance(image, torch.Tensor)
+            and not isinstance(image, PIL.Image.Image)
+            and not isinstance(image, list)
+        ):
+            raise ValueError(
+                "`image` has to be of type `torch.Tensor` or `PIL.Image.Image` or `List[PIL.Image.Image]` but is"
+                f" {type(image)}"
+            )
+
+    def prepare_image_latents(
+        self,
+        image,
+        device,
+        num_frames,
+        num_videos_per_prompt,
+    ):
+        image = image.to(device=device)
+        image_latents = self.vae.encode(image).latent_dist.sample()
+        image_latents = image_latents * self.vae.config.scaling_factor
+
+        # Add frames dimension to image latents
+        image_latents = image_latents.unsqueeze(2)
+
+        # Append a position mask for each subsequent frame
+        # after the initial image latent frame
+        frame_position_mask = []
+        for frame_idx in range(num_frames - 1):
+            scale = (frame_idx + 1) / (num_frames - 1)
+            frame_position_mask.append(torch.ones_like(image_latents[:, :, :1]) * scale)
+        if frame_position_mask:
+            frame_position_mask = torch.cat(frame_position_mask, dim=2)
+            image_latents = torch.cat([image_latents, frame_position_mask], dim=2)
+
+        # duplicate image_latents for each generation per prompt, using mps friendly method
+        image_latents = image_latents.repeat(num_videos_per_prompt, 1, 1, 1, 1)
+
+        if self.do_classifier_free_guidance:
+            image_latents = torch.cat([image_latents] * 2)
+
+        return image_latents
+
+    # Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_synth.TextToVideoSDPipeline.prepare_latents
+    def prepare_latents(
+        self, batch_size, num_channels_latents, num_frames, height, width, dtype, device, generator, latents=None
+    ):
+        shape = (
+            batch_size,
+            num_channels_latents,
+            num_frames,
+            height // self.vae_scale_factor,
+            width // self.vae_scale_factor,
+        )
+        if isinstance(generator, list) and len(generator) != batch_size:
+            raise ValueError(
+                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
+                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+            )
+
+        if latents is None:
+            latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        else:
+            latents = latents.to(device)
+
+        # scale the initial noise by the standard deviation required by the scheduler
+        latents = latents * self.scheduler.init_noise_sigma
+        return latents
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]] = None,
+        image: PipelineImageInput = None,
+        height: Optional[int] = 704,
+        width: Optional[int] = 1280,
+        target_fps: Optional[int] = 16,
+        num_frames: int = 16,
+        num_inference_steps: int = 50,
+        guidance_scale: float = 9.0,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        eta: float = 0.0,
+        num_videos_per_prompt: Optional[int] = 1,
+        decode_chunk_size: Optional[int] = 1,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        output_type: Optional[str] = "pil",
+        return_dict: bool = True,
+        cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+        clip_skip: Optional[int] = 1,
+    ):
+        r"""
+        The call function to the pipeline for image-to-video generation with [`I2VGenXLPipeline`].
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
+            image (`PIL.Image.Image` or `List[PIL.Image.Image]` or `torch.Tensor`):
+                Image or images to guide image generation. If you provide a tensor, it needs to be compatible with
+                [`CLIPImageProcessor`](https://huggingface.co/lambdalabs/sd-image-variations-diffusers/blob/main/feature_extractor/preprocessor_config.json).
+            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.
+            target_fps (`int`, *optional*):
+                Frames per second. The rate at which the generated images shall be exported to a video after
+                generation. This is also used as a "micro-condition" while generation.
+            num_frames (`int`, *optional*):
+                The number of video frames to generate.
+            num_inference_steps (`int`, *optional*):
+                The number of denoising steps.
+            guidance_scale (`float`, *optional*, defaults to 7.5):
+                A higher guidance scale value encourages the model to generate images closely linked to the text
+                `prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide what to not include in image generation. If not defined, you need to
+                pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`).
+            eta (`float`, *optional*):
+                Corresponds to parameter eta (η) from the [DDIM](https://huggingface.co/papers/2010.02502) paper. Only
+                applies to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers.
+            num_videos_per_prompt (`int`, *optional*):
+                The number of images to generate per prompt.
+            decode_chunk_size (`int`, *optional*):
+                The number of frames to decode at a time. The higher the chunk size, the higher the temporal
+                consistency between frames, but also the higher the memory consumption. By default, the decoder will
+                decode all frames at once for maximal quality. Reduce `decode_chunk_size` to reduce memory usage.
+            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.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 is generated by sampling using the supplied random `generator`.
+            prompt_embeds (`torch.Tensor`, *optional*):
+                Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
+                provided, text embeddings are generated from the `prompt` input argument.
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
+                Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
+                not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generated image. Choose between `PIL.Image` or `np.array`.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] instead of a
+                plain tuple.
+            cross_attention_kwargs (`dict`, *optional*):
+                A kwargs dictionary that if specified is passed along to the [`AttentionProcessor`] as defined in
+                [`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+            clip_skip (`int`, *optional*):
+                Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
+                the output of the pre-final layer will be used for computing the prompt embeddings.
+
+        Examples:
+
+        Returns:
+            [`pipelines.i2vgen_xl.pipeline_i2vgen_xl.I2VGenXLPipelineOutput`] or `tuple`:
+                If `return_dict` is `True`, [`pipelines.i2vgen_xl.pipeline_i2vgen_xl.I2VGenXLPipelineOutput`] is
+                returned, otherwise a `tuple` is returned where the first element is a list with the generated frames.
+        """
+        # 0. Default height and width to unet
+        height = height or self.unet.config.sample_size * self.vae_scale_factor
+        width = width or self.unet.config.sample_size * self.vae_scale_factor
+
+        # 1. Check inputs. Raise error if not correct
+        self.check_inputs(prompt, image, height, width, negative_prompt, prompt_embeds, negative_prompt_embeds)
+
+        # 2. Define call parameters
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        device = self._execution_device
+        # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
+        # of the Imagen paper: https://huggingface.co/papers/2205.11487 . `guidance_scale = 1`
+        # corresponds to doing no classifier free guidance.
+        self._guidance_scale = guidance_scale
+
+        # 3.1 Encode input text prompt
+        prompt_embeds, negative_prompt_embeds = self.encode_prompt(
+            prompt,
+            device,
+            num_videos_per_prompt,
+            negative_prompt,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
+            clip_skip=clip_skip,
+        )
+        # For classifier free guidance, we need to do two forward passes.
+        # Here we concatenate the unconditional and text embeddings into a single batch
+        # to avoid doing two forward passes
+        if self.do_classifier_free_guidance:
+            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
+
+        # 3.2 Encode image prompt
+        # 3.2.1 Image encodings.
+        # https://github.com/ali-vilab/i2vgen-xl/blob/2539c9262ff8a2a22fa9daecbfd13f0a2dbc32d0/tools/inferences/inference_i2vgen_entrance.py#L114
+        cropped_image = _center_crop_wide(image, (width, width))
+        cropped_image = _resize_bilinear(
+            cropped_image, (self.feature_extractor.crop_size["width"], self.feature_extractor.crop_size["height"])
+        )
+        image_embeddings = self._encode_image(cropped_image, device, num_videos_per_prompt)
+
+        # 3.2.2 Image latents.
+        resized_image = _center_crop_wide(image, (width, height))
+        image = self.video_processor.preprocess(resized_image).to(device=device, dtype=image_embeddings.dtype)
+        image_latents = self.prepare_image_latents(
+            image,
+            device=device,
+            num_frames=num_frames,
+            num_videos_per_prompt=num_videos_per_prompt,
+        )
+
+        # 3.3 Prepare additional conditions for the UNet.
+        if self.do_classifier_free_guidance:
+            fps_tensor = torch.tensor([target_fps, target_fps]).to(device)
+        else:
+            fps_tensor = torch.tensor([target_fps]).to(device)
+        fps_tensor = fps_tensor.repeat(batch_size * num_videos_per_prompt, 1).ravel()
+
+        # 4. Prepare timesteps
+        self.scheduler.set_timesteps(num_inference_steps, device=device)
+        timesteps = self.scheduler.timesteps
+
+        # 5. Prepare latent variables
+        num_channels_latents = self.unet.config.in_channels
+        latents = self.prepare_latents(
+            batch_size * num_videos_per_prompt,
+            num_channels_latents,
+            num_frames,
+            height,
+            width,
+            prompt_embeds.dtype,
+            device,
+            generator,
+            latents,
+        )
+
+        # 6. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
+        extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)
+
+        # 7. Denoising loop
+        num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order
+        with self.progress_bar(total=num_inference_steps) as progress_bar:
+            for i, t in enumerate(timesteps):
+                # expand the latents if we are doing classifier free guidance
+                latent_model_input = torch.cat([latents] * 2) if self.do_classifier_free_guidance else latents
+                latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
+
+                # predict the noise residual
+                noise_pred = self.unet(
+                    latent_model_input,
+                    t,
+                    encoder_hidden_states=prompt_embeds,
+                    fps=fps_tensor,
+                    image_latents=image_latents,
+                    image_embeddings=image_embeddings,
+                    cross_attention_kwargs=cross_attention_kwargs,
+                    return_dict=False,
+                )[0]
+
+                # perform guidance
+                if self.do_classifier_free_guidance:
+                    noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
+                    noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
+
+                # reshape latents
+                batch_size, channel, frames, width, height = latents.shape
+                latents = latents.permute(0, 2, 1, 3, 4).reshape(batch_size * frames, channel, width, height)
+                noise_pred = noise_pred.permute(0, 2, 1, 3, 4).reshape(batch_size * frames, channel, width, height)
+
+                # compute the previous noisy sample x_t -> x_t-1
+                latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs).prev_sample
+
+                # reshape latents back
+                latents = latents[None, :].reshape(batch_size, frames, channel, width, height).permute(0, 2, 1, 3, 4)
+                # call the callback, if provided
+                if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
+                    progress_bar.update()
+
+                if XLA_AVAILABLE:
+                    xm.mark_step()
+
+        # 8. Post processing
+        if output_type == "latent":
+            video = latents
+        else:
+            video_tensor = self.decode_latents(latents, decode_chunk_size=decode_chunk_size)
+            video = self.video_processor.postprocess_video(video=video_tensor, output_type=output_type)
+
+        # 9. Offload all models
+        self.maybe_free_model_hooks()
+
+        if not return_dict:
+            return (video,)
+
+        return I2VGenXLPipelineOutput(frames=video)
+
+
+# The following utilities are taken and adapted from
+# https://github.com/ali-vilab/i2vgen-xl/blob/main/utils/transforms.py.
+
+
+def _convert_pt_to_pil(image: Union[torch.Tensor, List[torch.Tensor]]):
+    if isinstance(image, list) and isinstance(image[0], torch.Tensor):
+        image = torch.cat(image, 0)
+
+    if isinstance(image, torch.Tensor):
+        if image.ndim == 3:
+            image = image.unsqueeze(0)
+
+        image_numpy = VaeImageProcessor.pt_to_numpy(image)
+        image_pil = VaeImageProcessor.numpy_to_pil(image_numpy)
+        image = image_pil
+
+    return image
+
+
+def _resize_bilinear(
+    image: Union[torch.Tensor, List[torch.Tensor], PIL.Image.Image, List[PIL.Image.Image]], resolution: Tuple[int, int]
+):
+    # First convert the images to PIL in case they are float tensors (only relevant for tests now).
+    image = _convert_pt_to_pil(image)
+
+    if isinstance(image, list):
+        image = [u.resize(resolution, PIL.Image.BILINEAR) for u in image]
+    else:
+        image = image.resize(resolution, PIL.Image.BILINEAR)
+    return image
+
+
+def _center_crop_wide(
+    image: Union[torch.Tensor, List[torch.Tensor], PIL.Image.Image, List[PIL.Image.Image]], resolution: Tuple[int, int]
+):
+    # First convert the images to PIL in case they are float tensors (only relevant for tests now).
+    image = _convert_pt_to_pil(image)
+
+    if isinstance(image, list):
+        scale = min(image[0].size[0] / resolution[0], image[0].size[1] / resolution[1])
+        image = [u.resize((round(u.width // scale), round(u.height // scale)), resample=PIL.Image.BOX) for u in image]
+
+        # center crop
+        x1 = (image[0].width - resolution[0]) // 2
+        y1 = (image[0].height - resolution[1]) // 2
+        image = [u.crop((x1, y1, x1 + resolution[0], y1 + resolution[1])) for u in image]
+        return image
+    else:
+        scale = min(image.size[0] / resolution[0], image.size[1] / resolution[1])
+        image = image.resize((round(image.width // scale), round(image.height // scale)), resample=PIL.Image.BOX)
+        x1 = (image.width - resolution[0]) // 2
+        y1 = (image.height - resolution[1]) // 2
+        image = image.crop((x1, y1, x1 + resolution[0], y1 + resolution[1]))
+        return image

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/kandinsky/__init__.py

@@ -0,0 +1,66 @@
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    DIFFUSERS_SLOW_IMPORT,
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    get_objects_from_module,
+    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_kandinsky"] = ["KandinskyPipeline"]
+    _import_structure["pipeline_kandinsky_combined"] = [
+        "KandinskyCombinedPipeline",
+        "KandinskyImg2ImgCombinedPipeline",
+        "KandinskyInpaintCombinedPipeline",
+    ]
+    _import_structure["pipeline_kandinsky_img2img"] = ["KandinskyImg2ImgPipeline"]
+    _import_structure["pipeline_kandinsky_inpaint"] = ["KandinskyInpaintPipeline"]
+    _import_structure["pipeline_kandinsky_prior"] = ["KandinskyPriorPipeline", "KandinskyPriorPipelineOutput"]
+    _import_structure["text_encoder"] = ["MultilingualCLIP"]
+
+
+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_kandinsky import KandinskyPipeline
+        from .pipeline_kandinsky_combined import (
+            KandinskyCombinedPipeline,
+            KandinskyImg2ImgCombinedPipeline,
+            KandinskyInpaintCombinedPipeline,
+        )
+        from .pipeline_kandinsky_img2img import KandinskyImg2ImgPipeline
+        from .pipeline_kandinsky_inpaint import KandinskyInpaintPipeline
+        from .pipeline_kandinsky_prior import KandinskyPriorPipeline, KandinskyPriorPipelineOutput
+        from .text_encoder import MultilingualCLIP
+
+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)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/kandinsky/pipeline_kandinsky.py

@@ -0,0 +1,419 @@
+# Copyright 2025 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.
+
+from typing import Callable, List, Optional, Union
+
+import torch
+from transformers import (
+    XLMRobertaTokenizer,
+)
+
+from ...models import UNet2DConditionModel, VQModel
+from ...schedulers import DDIMScheduler, DDPMScheduler
+from ...utils import (
+    is_torch_xla_available,
+    logging,
+    replace_example_docstring,
+)
+from ...utils.torch_utils import randn_tensor
+from ..pipeline_utils import DiffusionPipeline, ImagePipelineOutput
+from .text_encoder import MultilingualCLIP
+
+
+if is_torch_xla_available():
+    import torch_xla.core.xla_model as xm
+
+    XLA_AVAILABLE = True
+else:
+    XLA_AVAILABLE = False
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+
+EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> from diffusers import KandinskyPipeline, KandinskyPriorPipeline
+        >>> import torch
+
+        >>> pipe_prior = KandinskyPriorPipeline.from_pretrained("kandinsky-community/Kandinsky-2-1-prior")
+        >>> pipe_prior.to("cuda")
+
+        >>> prompt = "red cat, 4k photo"
+        >>> out = pipe_prior(prompt)
+        >>> image_emb = out.image_embeds
+        >>> negative_image_emb = out.negative_image_embeds
+
+        >>> pipe = KandinskyPipeline.from_pretrained("kandinsky-community/kandinsky-2-1")
+        >>> pipe.to("cuda")
+
+        >>> image = pipe(
+        ...     prompt,
+        ...     image_embeds=image_emb,
+        ...     negative_image_embeds=negative_image_emb,
+        ...     height=768,
+        ...     width=768,
+        ...     num_inference_steps=100,
+        ... ).images
+
+        >>> image[0].save("cat.png")
+        ```
+"""
+
+
+def get_new_h_w(h, w, scale_factor=8):
+    new_h = h // scale_factor**2
+    if h % scale_factor**2 != 0:
+        new_h += 1
+    new_w = w // scale_factor**2
+    if w % scale_factor**2 != 0:
+        new_w += 1
+    return new_h * scale_factor, new_w * scale_factor
+
+
+class KandinskyPipeline(DiffusionPipeline):
+    """
+    Pipeline for text-to-image generation using Kandinsky
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
+    library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
+
+    Args:
+        text_encoder ([`MultilingualCLIP`]):
+            Frozen text-encoder.
+        tokenizer ([`XLMRobertaTokenizer`]):
+            Tokenizer of class
+        scheduler (Union[`DDIMScheduler`,`DDPMScheduler`]):
+            A scheduler to be used in combination with `unet` to generate image latents.
+        unet ([`UNet2DConditionModel`]):
+            Conditional U-Net architecture to denoise the image embedding.
+        movq ([`VQModel`]):
+            MoVQ Decoder to generate the image from the latents.
+    """
+
+    model_cpu_offload_seq = "text_encoder->unet->movq"
+
+    def __init__(
+        self,
+        text_encoder: MultilingualCLIP,
+        tokenizer: XLMRobertaTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: Union[DDIMScheduler, DDPMScheduler],
+        movq: VQModel,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            movq=movq,
+        )
+        self.movq_scale_factor = 2 ** (len(self.movq.config.block_out_channels) - 1)
+
+    # Copied from diffusers.pipelines.unclip.pipeline_unclip.UnCLIPPipeline.prepare_latents
+    def prepare_latents(self, shape, dtype, device, generator, latents, scheduler):
+        if latents is None:
+            latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        else:
+            if latents.shape != shape:
+                raise ValueError(f"Unexpected latents shape, got {latents.shape}, expected {shape}")
+            latents = latents.to(device)
+
+        latents = latents * scheduler.init_noise_sigma
+        return latents
+
+    def _encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+    ):
+        batch_size = len(prompt) if isinstance(prompt, list) else 1
+        # get prompt text embeddings
+        text_inputs = self.tokenizer(
+            prompt,
+            padding="max_length",
+            truncation=True,
+            max_length=77,
+            return_attention_mask=True,
+            add_special_tokens=True,
+            return_tensors="pt",
+        )
+
+        text_input_ids = text_inputs.input_ids
+        untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+        if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(text_input_ids, untruncated_ids):
+            removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1])
+            logger.warning(
+                "The following part of your input was truncated because CLIP can only handle sequences up to"
+                f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+            )
+
+        text_input_ids = text_input_ids.to(device)
+        text_mask = text_inputs.attention_mask.to(device)
+
+        prompt_embeds, text_encoder_hidden_states = self.text_encoder(
+            input_ids=text_input_ids, attention_mask=text_mask
+        )
+
+        prompt_embeds = prompt_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+        text_encoder_hidden_states = text_encoder_hidden_states.repeat_interleave(num_images_per_prompt, dim=0)
+        text_mask = text_mask.repeat_interleave(num_images_per_prompt, dim=0)
+
+        if do_classifier_free_guidance:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=77,
+                truncation=True,
+                return_attention_mask=True,
+                add_special_tokens=True,
+                return_tensors="pt",
+            )
+            uncond_text_input_ids = uncond_input.input_ids.to(device)
+            uncond_text_mask = uncond_input.attention_mask.to(device)
+
+            negative_prompt_embeds, uncond_text_encoder_hidden_states = self.text_encoder(
+                input_ids=uncond_text_input_ids, attention_mask=uncond_text_mask
+            )
+
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+
+            seq_len = negative_prompt_embeds.shape[1]
+            negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt)
+            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len)
+
+            seq_len = uncond_text_encoder_hidden_states.shape[1]
+            uncond_text_encoder_hidden_states = uncond_text_encoder_hidden_states.repeat(1, num_images_per_prompt, 1)
+            uncond_text_encoder_hidden_states = uncond_text_encoder_hidden_states.view(
+                batch_size * num_images_per_prompt, seq_len, -1
+            )
+            uncond_text_mask = uncond_text_mask.repeat_interleave(num_images_per_prompt, dim=0)
+
+            # done duplicates
+
+            # For classifier free guidance, we need to do two forward passes.
+            # Here we concatenate the unconditional and text embeddings into a single batch
+            # to avoid doing two forward passes
+            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
+            text_encoder_hidden_states = torch.cat([uncond_text_encoder_hidden_states, text_encoder_hidden_states])
+
+            text_mask = torch.cat([uncond_text_mask, text_mask])
+
+        return prompt_embeds, text_encoder_hidden_states, text_mask
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        image_embeds: Union[torch.Tensor, List[torch.Tensor]],
+        negative_image_embeds: Union[torch.Tensor, List[torch.Tensor]],
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        height: int = 512,
+        width: int = 512,
+        num_inference_steps: int = 100,
+        guidance_scale: float = 4.0,
+        num_images_per_prompt: int = 1,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: Optional[str] = "pil",
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
+        callback_steps: int = 1,
+        return_dict: bool = True,
+    ):
+        """
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`):
+                The prompt or prompts to guide the image generation.
+            image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
+                The clip image embeddings for text prompt, that will be used to condition the image generation.
+            negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
+                The clip image embeddings for negative text prompt, will be used to condition the image generation.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
+                if `guidance_scale` is less than `1`).
+            height (`int`, *optional*, defaults to 512):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to 512):
+                The width in pixels of the generated image.
+            num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            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.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 be generated by sampling using the supplied random `generator`.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generate image. Choose between: `"pil"` (`PIL.Image.Image`), `"np"`
+                (`np.array`) or `"pt"` (`torch.Tensor`).
+            callback (`Callable`, *optional*):
+                A function that calls every `callback_steps` steps during inference. The function is 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 is called. If not specified, the callback is called at
+                every step.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
+
+        Examples:
+
+        Returns:
+            [`~pipelines.ImagePipelineOutput`] or `tuple`
+        """
+
+        if isinstance(prompt, str):
+            batch_size = 1
+        elif isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        device = self._execution_device
+
+        batch_size = batch_size * num_images_per_prompt
+        do_classifier_free_guidance = guidance_scale > 1.0
+
+        prompt_embeds, text_encoder_hidden_states, _ = self._encode_prompt(
+            prompt, device, num_images_per_prompt, do_classifier_free_guidance, negative_prompt
+        )
+
+        if isinstance(image_embeds, list):
+            image_embeds = torch.cat(image_embeds, dim=0)
+        if isinstance(negative_image_embeds, list):
+            negative_image_embeds = torch.cat(negative_image_embeds, dim=0)
+
+        if do_classifier_free_guidance:
+            image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+            negative_image_embeds = negative_image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+
+            image_embeds = torch.cat([negative_image_embeds, image_embeds], dim=0).to(
+                dtype=prompt_embeds.dtype, device=device
+            )
+
+        self.scheduler.set_timesteps(num_inference_steps, device=device)
+        timesteps_tensor = self.scheduler.timesteps
+
+        num_channels_latents = self.unet.config.in_channels
+
+        height, width = get_new_h_w(height, width, self.movq_scale_factor)
+
+        # create initial latent
+        latents = self.prepare_latents(
+            (batch_size, num_channels_latents, height, width),
+            text_encoder_hidden_states.dtype,
+            device,
+            generator,
+            latents,
+            self.scheduler,
+        )
+
+        for i, t in enumerate(self.progress_bar(timesteps_tensor)):
+            # expand the latents if we are doing classifier free guidance
+            latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
+
+            added_cond_kwargs = {"text_embeds": prompt_embeds, "image_embeds": image_embeds}
+            noise_pred = self.unet(
+                sample=latent_model_input,
+                timestep=t,
+                encoder_hidden_states=text_encoder_hidden_states,
+                added_cond_kwargs=added_cond_kwargs,
+                return_dict=False,
+            )[0]
+
+            if do_classifier_free_guidance:
+                noise_pred, variance_pred = noise_pred.split(latents.shape[1], dim=1)
+                noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
+                _, variance_pred_text = variance_pred.chunk(2)
+                noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
+                noise_pred = torch.cat([noise_pred, variance_pred_text], dim=1)
+
+            if not (
+                hasattr(self.scheduler.config, "variance_type")
+                and self.scheduler.config.variance_type in ["learned", "learned_range"]
+            ):
+                noise_pred, _ = noise_pred.split(latents.shape[1], dim=1)
+
+            # compute the previous noisy sample x_t -> x_t-1
+            latents = self.scheduler.step(
+                noise_pred,
+                t,
+                latents,
+                generator=generator,
+            ).prev_sample
+
+            if callback is not None and i % callback_steps == 0:
+                step_idx = i // getattr(self.scheduler, "order", 1)
+                callback(step_idx, t, latents)
+
+            if XLA_AVAILABLE:
+                xm.mark_step()
+
+        # post-processing
+        image = self.movq.decode(latents, force_not_quantize=True)["sample"]
+
+        self.maybe_free_model_hooks()
+
+        if output_type not in ["pt", "np", "pil"]:
+            raise ValueError(f"Only the output types `pt`, `pil` and `np` are supported not output_type={output_type}")
+
+        if output_type in ["np", "pil"]:
+            image = image * 0.5 + 0.5
+            image = image.clamp(0, 1)
+            image = image.cpu().permute(0, 2, 3, 1).float().numpy()
+
+        if output_type == "pil":
+            image = self.numpy_to_pil(image)
+
+        if not return_dict:
+            return (image,)
+
+        return ImagePipelineOutput(images=image)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/kandinsky/pipeline_kandinsky_combined.py

@@ -0,0 +1,817 @@
+# Copyright 2025 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.
+from typing import Callable, List, Optional, Union
+
+import PIL.Image
+import torch
+from transformers import (
+    CLIPImageProcessor,
+    CLIPTextModelWithProjection,
+    CLIPTokenizer,
+    CLIPVisionModelWithProjection,
+    XLMRobertaTokenizer,
+)
+
+from ...models import PriorTransformer, UNet2DConditionModel, VQModel
+from ...schedulers import DDIMScheduler, DDPMScheduler, UnCLIPScheduler
+from ...utils import (
+    replace_example_docstring,
+)
+from ..pipeline_utils import DiffusionPipeline
+from .pipeline_kandinsky import KandinskyPipeline
+from .pipeline_kandinsky_img2img import KandinskyImg2ImgPipeline
+from .pipeline_kandinsky_inpaint import KandinskyInpaintPipeline
+from .pipeline_kandinsky_prior import KandinskyPriorPipeline
+from .text_encoder import MultilingualCLIP
+
+
+TEXT2IMAGE_EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        from diffusers import AutoPipelineForText2Image
+        import torch
+
+        pipe = AutoPipelineForText2Image.from_pretrained(
+            "kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16
+        )
+        pipe.enable_model_cpu_offload()
+
+        prompt = "A lion in galaxies, spirals, nebulae, stars, smoke, iridescent, intricate detail, octane render, 8k"
+
+        image = pipe(prompt=prompt, num_inference_steps=25).images[0]
+        ```
+"""
+
+IMAGE2IMAGE_EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        from diffusers import AutoPipelineForImage2Image
+        import torch
+        import requests
+        from io import BytesIO
+        from PIL import Image
+        import os
+
+        pipe = AutoPipelineForImage2Image.from_pretrained(
+            "kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16
+        )
+        pipe.enable_model_cpu_offload()
+
+        prompt = "A fantasy landscape, Cinematic lighting"
+        negative_prompt = "low quality, bad quality"
+
+        url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
+
+        response = requests.get(url)
+        image = Image.open(BytesIO(response.content)).convert("RGB")
+        image.thumbnail((768, 768))
+
+        image = pipe(prompt=prompt, image=original_image, num_inference_steps=25).images[0]
+        ```
+"""
+
+INPAINT_EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        from diffusers import AutoPipelineForInpainting
+        from diffusers.utils import load_image
+        import torch
+        import numpy as np
+
+        pipe = AutoPipelineForInpainting.from_pretrained(
+            "kandinsky-community/kandinsky-2-1-inpaint", torch_dtype=torch.float16
+        )
+        pipe.enable_model_cpu_offload()
+
+        prompt = "A fantasy landscape, Cinematic lighting"
+        negative_prompt = "low quality, bad quality"
+
+        original_image = load_image(
+            "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinsky/cat.png"
+        )
+
+        mask = np.zeros((768, 768), dtype=np.float32)
+        # Let's mask out an area above the cat's head
+        mask[:250, 250:-250] = 1
+
+        image = pipe(prompt=prompt, image=original_image, mask_image=mask, num_inference_steps=25).images[0]
+        ```
+"""
+
+
+class KandinskyCombinedPipeline(DiffusionPipeline):
+    """
+    Combined Pipeline for text-to-image generation using Kandinsky
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
+    library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
+
+    Args:
+        text_encoder ([`MultilingualCLIP`]):
+            Frozen text-encoder.
+        tokenizer ([`XLMRobertaTokenizer`]):
+            Tokenizer of class
+        scheduler (Union[`DDIMScheduler`,`DDPMScheduler`]):
+            A scheduler to be used in combination with `unet` to generate image latents.
+        unet ([`UNet2DConditionModel`]):
+            Conditional U-Net architecture to denoise the image embedding.
+        movq ([`VQModel`]):
+            MoVQ Decoder to generate the image from the latents.
+        prior_prior ([`PriorTransformer`]):
+            The canonical unCLIP prior to approximate the image embedding from the text embedding.
+        prior_image_encoder ([`CLIPVisionModelWithProjection`]):
+            Frozen image-encoder.
+        prior_text_encoder ([`CLIPTextModelWithProjection`]):
+            Frozen text-encoder.
+        prior_tokenizer (`CLIPTokenizer`):
+             Tokenizer of class
+             [CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer).
+        prior_scheduler ([`UnCLIPScheduler`]):
+            A scheduler to be used in combination with `prior` to generate image embedding.
+    """
+
+    _load_connected_pipes = True
+    model_cpu_offload_seq = "text_encoder->unet->movq->prior_prior->prior_image_encoder->prior_text_encoder"
+    _exclude_from_cpu_offload = ["prior_prior"]
+
+    def __init__(
+        self,
+        text_encoder: MultilingualCLIP,
+        tokenizer: XLMRobertaTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: Union[DDIMScheduler, DDPMScheduler],
+        movq: VQModel,
+        prior_prior: PriorTransformer,
+        prior_image_encoder: CLIPVisionModelWithProjection,
+        prior_text_encoder: CLIPTextModelWithProjection,
+        prior_tokenizer: CLIPTokenizer,
+        prior_scheduler: UnCLIPScheduler,
+        prior_image_processor: CLIPImageProcessor,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            movq=movq,
+            prior_prior=prior_prior,
+            prior_image_encoder=prior_image_encoder,
+            prior_text_encoder=prior_text_encoder,
+            prior_tokenizer=prior_tokenizer,
+            prior_scheduler=prior_scheduler,
+            prior_image_processor=prior_image_processor,
+        )
+        self.prior_pipe = KandinskyPriorPipeline(
+            prior=prior_prior,
+            image_encoder=prior_image_encoder,
+            text_encoder=prior_text_encoder,
+            tokenizer=prior_tokenizer,
+            scheduler=prior_scheduler,
+            image_processor=prior_image_processor,
+        )
+        self.decoder_pipe = KandinskyPipeline(
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            movq=movq,
+        )
+
+    def enable_xformers_memory_efficient_attention(self, attention_op: Optional[Callable] = None):
+        self.decoder_pipe.enable_xformers_memory_efficient_attention(attention_op)
+
+    def enable_sequential_cpu_offload(self, gpu_id: Optional[int] = None, device: Union[torch.device, str] = None):
+        r"""
+        Offloads all models (`unet`, `text_encoder`, `vae`, and `safety checker` state dicts) to CPU using 🤗
+        Accelerate, significantly reducing memory usage. Models are moved to a `torch.device('meta')` and loaded on a
+        GPU only when their specific submodule's `forward` method is called. Offloading happens on a submodule basis.
+        Memory savings are higher than using `enable_model_cpu_offload`, but performance is lower.
+        """
+        self.prior_pipe.enable_sequential_cpu_offload(gpu_id=gpu_id, device=device)
+        self.decoder_pipe.enable_sequential_cpu_offload(gpu_id=gpu_id, device=device)
+
+    def progress_bar(self, iterable=None, total=None):
+        self.prior_pipe.progress_bar(iterable=iterable, total=total)
+        self.decoder_pipe.progress_bar(iterable=iterable, total=total)
+        self.decoder_pipe.enable_model_cpu_offload()
+
+    def set_progress_bar_config(self, **kwargs):
+        self.prior_pipe.set_progress_bar_config(**kwargs)
+        self.decoder_pipe.set_progress_bar_config(**kwargs)
+
+    @torch.no_grad()
+    @replace_example_docstring(TEXT2IMAGE_EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        num_inference_steps: int = 100,
+        guidance_scale: float = 4.0,
+        num_images_per_prompt: int = 1,
+        height: int = 512,
+        width: int = 512,
+        prior_guidance_scale: float = 4.0,
+        prior_num_inference_steps: int = 25,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: Optional[str] = "pil",
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
+        callback_steps: int = 1,
+        return_dict: bool = True,
+    ):
+        """
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`):
+                The prompt or prompts to guide the image generation.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
+                if `guidance_scale` is less than `1`).
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            height (`int`, *optional*, defaults to 512):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to 512):
+                The width in pixels of the generated image.
+            prior_guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            prior_num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            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.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 be generated by sampling using the supplied random `generator`.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generate image. Choose between: `"pil"` (`PIL.Image.Image`), `"np"`
+                (`np.array`) or `"pt"` (`torch.Tensor`).
+            callback (`Callable`, *optional*):
+                A function that calls every `callback_steps` steps during inference. The function is 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 is called. If not specified, the callback is called at
+                every step.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
+
+        Examples:
+
+        Returns:
+            [`~pipelines.ImagePipelineOutput`] or `tuple`
+        """
+        prior_outputs = self.prior_pipe(
+            prompt=prompt,
+            negative_prompt=negative_prompt,
+            num_images_per_prompt=num_images_per_prompt,
+            num_inference_steps=prior_num_inference_steps,
+            generator=generator,
+            latents=latents,
+            guidance_scale=prior_guidance_scale,
+            output_type="pt",
+            return_dict=False,
+        )
+        image_embeds = prior_outputs[0]
+        negative_image_embeds = prior_outputs[1]
+
+        prompt = [prompt] if not isinstance(prompt, (list, tuple)) else prompt
+
+        if len(prompt) < image_embeds.shape[0] and image_embeds.shape[0] % len(prompt) == 0:
+            prompt = (image_embeds.shape[0] // len(prompt)) * prompt
+
+        outputs = self.decoder_pipe(
+            prompt=prompt,
+            image_embeds=image_embeds,
+            negative_image_embeds=negative_image_embeds,
+            width=width,
+            height=height,
+            num_inference_steps=num_inference_steps,
+            generator=generator,
+            guidance_scale=guidance_scale,
+            output_type=output_type,
+            callback=callback,
+            callback_steps=callback_steps,
+            return_dict=return_dict,
+        )
+
+        self.maybe_free_model_hooks()
+
+        return outputs
+
+
+class KandinskyImg2ImgCombinedPipeline(DiffusionPipeline):
+    """
+    Combined Pipeline for image-to-image generation using Kandinsky
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
+    library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
+
+    Args:
+        text_encoder ([`MultilingualCLIP`]):
+            Frozen text-encoder.
+        tokenizer ([`XLMRobertaTokenizer`]):
+            Tokenizer of class
+        scheduler (Union[`DDIMScheduler`,`DDPMScheduler`]):
+            A scheduler to be used in combination with `unet` to generate image latents.
+        unet ([`UNet2DConditionModel`]):
+            Conditional U-Net architecture to denoise the image embedding.
+        movq ([`VQModel`]):
+            MoVQ Decoder to generate the image from the latents.
+        prior_prior ([`PriorTransformer`]):
+            The canonical unCLIP prior to approximate the image embedding from the text embedding.
+        prior_image_encoder ([`CLIPVisionModelWithProjection`]):
+            Frozen image-encoder.
+        prior_text_encoder ([`CLIPTextModelWithProjection`]):
+            Frozen text-encoder.
+        prior_tokenizer (`CLIPTokenizer`):
+             Tokenizer of class
+             [CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer).
+        prior_scheduler ([`UnCLIPScheduler`]):
+            A scheduler to be used in combination with `prior` to generate image embedding.
+    """
+
+    _load_connected_pipes = True
+    model_cpu_offload_seq = "prior_text_encoder->prior_image_encoder->prior_prior->text_encoder->unet->movq"
+    _exclude_from_cpu_offload = ["prior_prior"]
+
+    def __init__(
+        self,
+        text_encoder: MultilingualCLIP,
+        tokenizer: XLMRobertaTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: Union[DDIMScheduler, DDPMScheduler],
+        movq: VQModel,
+        prior_prior: PriorTransformer,
+        prior_image_encoder: CLIPVisionModelWithProjection,
+        prior_text_encoder: CLIPTextModelWithProjection,
+        prior_tokenizer: CLIPTokenizer,
+        prior_scheduler: UnCLIPScheduler,
+        prior_image_processor: CLIPImageProcessor,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            movq=movq,
+            prior_prior=prior_prior,
+            prior_image_encoder=prior_image_encoder,
+            prior_text_encoder=prior_text_encoder,
+            prior_tokenizer=prior_tokenizer,
+            prior_scheduler=prior_scheduler,
+            prior_image_processor=prior_image_processor,
+        )
+        self.prior_pipe = KandinskyPriorPipeline(
+            prior=prior_prior,
+            image_encoder=prior_image_encoder,
+            text_encoder=prior_text_encoder,
+            tokenizer=prior_tokenizer,
+            scheduler=prior_scheduler,
+            image_processor=prior_image_processor,
+        )
+        self.decoder_pipe = KandinskyImg2ImgPipeline(
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            movq=movq,
+        )
+
+    def enable_xformers_memory_efficient_attention(self, attention_op: Optional[Callable] = None):
+        self.decoder_pipe.enable_xformers_memory_efficient_attention(attention_op)
+
+    def enable_sequential_cpu_offload(self, gpu_id: Optional[int] = None, device: Union[torch.device, str] = None):
+        r"""
+        Offloads all models to CPU using accelerate, significantly reducing memory usage. When called, unet,
+        text_encoder, vae and safety checker have their state dicts saved to CPU and then are moved to a
+        `torch.device('meta') and loaded to GPU only when their specific submodule has its `forward` method called.
+        Note that offloading happens on a submodule basis. Memory savings are higher than with
+        `enable_model_cpu_offload`, but performance is lower.
+        """
+        self.prior_pipe.enable_sequential_cpu_offload(gpu_id=gpu_id, device=device)
+        self.decoder_pipe.enable_sequential_cpu_offload(gpu_id=gpu_id, device=device)
+
+    def progress_bar(self, iterable=None, total=None):
+        self.prior_pipe.progress_bar(iterable=iterable, total=total)
+        self.decoder_pipe.progress_bar(iterable=iterable, total=total)
+        self.decoder_pipe.enable_model_cpu_offload()
+
+    def set_progress_bar_config(self, **kwargs):
+        self.prior_pipe.set_progress_bar_config(**kwargs)
+        self.decoder_pipe.set_progress_bar_config(**kwargs)
+
+    @torch.no_grad()
+    @replace_example_docstring(IMAGE2IMAGE_EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        num_inference_steps: int = 100,
+        guidance_scale: float = 4.0,
+        num_images_per_prompt: int = 1,
+        strength: float = 0.3,
+        height: int = 512,
+        width: int = 512,
+        prior_guidance_scale: float = 4.0,
+        prior_num_inference_steps: int = 25,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: Optional[str] = "pil",
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
+        callback_steps: int = 1,
+        return_dict: bool = True,
+    ):
+        """
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`):
+                The prompt or prompts to guide the image generation.
+            image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`):
+                `Image`, or tensor representing an image batch, that will be used as the starting point for the
+                process. Can also accept image latents as `image`, if passing latents directly, it will not be encoded
+                again.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
+                if `guidance_scale` is less than `1`).
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            height (`int`, *optional*, defaults to 512):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to 512):
+                The width in pixels of the generated image.
+            strength (`float`, *optional*, defaults to 0.3):
+                Conceptually, indicates how much to transform the reference `image`. Must be between 0 and 1. `image`
+                will be used as a starting point, adding more noise to it the larger the `strength`. The number of
+                denoising steps depends on the amount of noise initially added. When `strength` is 1, added noise will
+                be maximum and the denoising process will run for the full number of iterations specified in
+                `num_inference_steps`. A value of 1, therefore, essentially ignores `image`.
+            prior_guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            prior_num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            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.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 be generated by sampling using the supplied random `generator`.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generate image. Choose between: `"pil"` (`PIL.Image.Image`), `"np"`
+                (`np.array`) or `"pt"` (`torch.Tensor`).
+            callback (`Callable`, *optional*):
+                A function that calls every `callback_steps` steps during inference. The function is 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 is called. If not specified, the callback is called at
+                every step.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
+
+        Examples:
+
+        Returns:
+            [`~pipelines.ImagePipelineOutput`] or `tuple`
+        """
+        prior_outputs = self.prior_pipe(
+            prompt=prompt,
+            negative_prompt=negative_prompt,
+            num_images_per_prompt=num_images_per_prompt,
+            num_inference_steps=prior_num_inference_steps,
+            generator=generator,
+            latents=latents,
+            guidance_scale=prior_guidance_scale,
+            output_type="pt",
+            return_dict=False,
+        )
+        image_embeds = prior_outputs[0]
+        negative_image_embeds = prior_outputs[1]
+
+        prompt = [prompt] if not isinstance(prompt, (list, tuple)) else prompt
+        image = [image] if isinstance(prompt, PIL.Image.Image) else image
+
+        if len(prompt) < image_embeds.shape[0] and image_embeds.shape[0] % len(prompt) == 0:
+            prompt = (image_embeds.shape[0] // len(prompt)) * prompt
+
+        if (
+            isinstance(image, (list, tuple))
+            and len(image) < image_embeds.shape[0]
+            and image_embeds.shape[0] % len(image) == 0
+        ):
+            image = (image_embeds.shape[0] // len(image)) * image
+
+        outputs = self.decoder_pipe(
+            prompt=prompt,
+            image=image,
+            image_embeds=image_embeds,
+            negative_image_embeds=negative_image_embeds,
+            strength=strength,
+            width=width,
+            height=height,
+            num_inference_steps=num_inference_steps,
+            generator=generator,
+            guidance_scale=guidance_scale,
+            output_type=output_type,
+            callback=callback,
+            callback_steps=callback_steps,
+            return_dict=return_dict,
+        )
+
+        self.maybe_free_model_hooks()
+
+        return outputs
+
+
+class KandinskyInpaintCombinedPipeline(DiffusionPipeline):
+    """
+    Combined Pipeline for generation using Kandinsky
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
+    library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
+
+    Args:
+        text_encoder ([`MultilingualCLIP`]):
+            Frozen text-encoder.
+        tokenizer ([`XLMRobertaTokenizer`]):
+            Tokenizer of class
+        scheduler (Union[`DDIMScheduler`,`DDPMScheduler`]):
+            A scheduler to be used in combination with `unet` to generate image latents.
+        unet ([`UNet2DConditionModel`]):
+            Conditional U-Net architecture to denoise the image embedding.
+        movq ([`VQModel`]):
+            MoVQ Decoder to generate the image from the latents.
+        prior_prior ([`PriorTransformer`]):
+            The canonical unCLIP prior to approximate the image embedding from the text embedding.
+        prior_image_encoder ([`CLIPVisionModelWithProjection`]):
+            Frozen image-encoder.
+        prior_text_encoder ([`CLIPTextModelWithProjection`]):
+            Frozen text-encoder.
+        prior_tokenizer (`CLIPTokenizer`):
+             Tokenizer of class
+             [CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer).
+        prior_scheduler ([`UnCLIPScheduler`]):
+            A scheduler to be used in combination with `prior` to generate image embedding.
+    """
+
+    _load_connected_pipes = True
+    model_cpu_offload_seq = "prior_text_encoder->prior_image_encoder->prior_prior->text_encoder->unet->movq"
+    _exclude_from_cpu_offload = ["prior_prior"]
+
+    def __init__(
+        self,
+        text_encoder: MultilingualCLIP,
+        tokenizer: XLMRobertaTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: Union[DDIMScheduler, DDPMScheduler],
+        movq: VQModel,
+        prior_prior: PriorTransformer,
+        prior_image_encoder: CLIPVisionModelWithProjection,
+        prior_text_encoder: CLIPTextModelWithProjection,
+        prior_tokenizer: CLIPTokenizer,
+        prior_scheduler: UnCLIPScheduler,
+        prior_image_processor: CLIPImageProcessor,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            movq=movq,
+            prior_prior=prior_prior,
+            prior_image_encoder=prior_image_encoder,
+            prior_text_encoder=prior_text_encoder,
+            prior_tokenizer=prior_tokenizer,
+            prior_scheduler=prior_scheduler,
+            prior_image_processor=prior_image_processor,
+        )
+        self.prior_pipe = KandinskyPriorPipeline(
+            prior=prior_prior,
+            image_encoder=prior_image_encoder,
+            text_encoder=prior_text_encoder,
+            tokenizer=prior_tokenizer,
+            scheduler=prior_scheduler,
+            image_processor=prior_image_processor,
+        )
+        self.decoder_pipe = KandinskyInpaintPipeline(
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            movq=movq,
+        )
+
+    def enable_xformers_memory_efficient_attention(self, attention_op: Optional[Callable] = None):
+        self.decoder_pipe.enable_xformers_memory_efficient_attention(attention_op)
+
+    def enable_sequential_cpu_offload(self, gpu_id: Optional[int] = None, device: Union[torch.device, str] = None):
+        r"""
+        Offloads all models to CPU using accelerate, significantly reducing memory usage. When called, unet,
+        text_encoder, vae and safety checker have their state dicts saved to CPU and then are moved to a
+        `torch.device('meta') and loaded to GPU only when their specific submodule has its `forward` method called.
+        Note that offloading happens on a submodule basis. Memory savings are higher than with
+        `enable_model_cpu_offload`, but performance is lower.
+        """
+        self.prior_pipe.enable_sequential_cpu_offload(gpu_id=gpu_id, device=device)
+        self.decoder_pipe.enable_sequential_cpu_offload(gpu_id=gpu_id, device=device)
+
+    def progress_bar(self, iterable=None, total=None):
+        self.prior_pipe.progress_bar(iterable=iterable, total=total)
+        self.decoder_pipe.progress_bar(iterable=iterable, total=total)
+        self.decoder_pipe.enable_model_cpu_offload()
+
+    def set_progress_bar_config(self, **kwargs):
+        self.prior_pipe.set_progress_bar_config(**kwargs)
+        self.decoder_pipe.set_progress_bar_config(**kwargs)
+
+    @torch.no_grad()
+    @replace_example_docstring(INPAINT_EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
+        mask_image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        num_inference_steps: int = 100,
+        guidance_scale: float = 4.0,
+        num_images_per_prompt: int = 1,
+        height: int = 512,
+        width: int = 512,
+        prior_guidance_scale: float = 4.0,
+        prior_num_inference_steps: int = 25,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: Optional[str] = "pil",
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
+        callback_steps: int = 1,
+        return_dict: bool = True,
+    ):
+        """
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`):
+                The prompt or prompts to guide the image generation.
+            image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`):
+                `Image`, or tensor representing an image batch, that will be used as the starting point for the
+                process. Can also accept image latents as `image`, if passing latents directly, it will not be encoded
+                again.
+            mask_image (`np.array`):
+                Tensor representing an image batch, to mask `image`. White pixels in the mask will be repainted, while
+                black pixels will be preserved. If `mask_image` is a PIL image, it will be converted to a single
+                channel (luminance) before use. If it's a tensor, it should contain one color channel (L) instead of 3,
+                so the expected shape would be `(B, H, W, 1)`.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
+                if `guidance_scale` is less than `1`).
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            height (`int`, *optional*, defaults to 512):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to 512):
+                The width in pixels of the generated image.
+            prior_guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            prior_num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            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.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 be generated by sampling using the supplied random `generator`.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generate image. Choose between: `"pil"` (`PIL.Image.Image`), `"np"`
+                (`np.array`) or `"pt"` (`torch.Tensor`).
+            callback (`Callable`, *optional*):
+                A function that calls every `callback_steps` steps during inference. The function is 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 is called. If not specified, the callback is called at
+                every step.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
+
+        Examples:
+
+        Returns:
+            [`~pipelines.ImagePipelineOutput`] or `tuple`
+        """
+        prior_outputs = self.prior_pipe(
+            prompt=prompt,
+            negative_prompt=negative_prompt,
+            num_images_per_prompt=num_images_per_prompt,
+            num_inference_steps=prior_num_inference_steps,
+            generator=generator,
+            latents=latents,
+            guidance_scale=prior_guidance_scale,
+            output_type="pt",
+            return_dict=False,
+        )
+        image_embeds = prior_outputs[0]
+        negative_image_embeds = prior_outputs[1]
+
+        prompt = [prompt] if not isinstance(prompt, (list, tuple)) else prompt
+        image = [image] if isinstance(prompt, PIL.Image.Image) else image
+        mask_image = [mask_image] if isinstance(mask_image, PIL.Image.Image) else mask_image
+
+        if len(prompt) < image_embeds.shape[0] and image_embeds.shape[0] % len(prompt) == 0:
+            prompt = (image_embeds.shape[0] // len(prompt)) * prompt
+
+        if (
+            isinstance(image, (list, tuple))
+            and len(image) < image_embeds.shape[0]
+            and image_embeds.shape[0] % len(image) == 0
+        ):
+            image = (image_embeds.shape[0] // len(image)) * image
+
+        if (
+            isinstance(mask_image, (list, tuple))
+            and len(mask_image) < image_embeds.shape[0]
+            and image_embeds.shape[0] % len(mask_image) == 0
+        ):
+            mask_image = (image_embeds.shape[0] // len(mask_image)) * mask_image
+
+        outputs = self.decoder_pipe(
+            prompt=prompt,
+            image=image,
+            mask_image=mask_image,
+            image_embeds=image_embeds,
+            negative_image_embeds=negative_image_embeds,
+            width=width,
+            height=height,
+            num_inference_steps=num_inference_steps,
+            generator=generator,
+            guidance_scale=guidance_scale,
+            output_type=output_type,
+            callback=callback,
+            callback_steps=callback_steps,
+            return_dict=return_dict,
+        )
+
+        self.maybe_free_model_hooks()
+
+        return outputs

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/kandinsky/pipeline_kandinsky_img2img.py

@@ -0,0 +1,505 @@
+# Copyright 2025 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.
+
+from typing import Callable, List, Optional, Union
+
+import PIL.Image
+import torch
+from transformers import (
+    XLMRobertaTokenizer,
+)
+
+from ...image_processor import VaeImageProcessor
+from ...models import UNet2DConditionModel, VQModel
+from ...schedulers import DDIMScheduler
+from ...utils import (
+    is_torch_xla_available,
+    logging,
+    replace_example_docstring,
+)
+from ...utils.torch_utils import randn_tensor
+from ..pipeline_utils import DiffusionPipeline, ImagePipelineOutput
+from .text_encoder import MultilingualCLIP
+
+
+if is_torch_xla_available():
+    import torch_xla.core.xla_model as xm
+
+    XLA_AVAILABLE = True
+else:
+    XLA_AVAILABLE = False
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+
+EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> from diffusers import KandinskyImg2ImgPipeline, KandinskyPriorPipeline
+        >>> from diffusers.utils import load_image
+        >>> import torch
+
+        >>> pipe_prior = KandinskyPriorPipeline.from_pretrained(
+        ...     "kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16
+        ... )
+        >>> pipe_prior.to("cuda")
+
+        >>> prompt = "A red cartoon frog, 4k"
+        >>> image_emb, zero_image_emb = pipe_prior(prompt, return_dict=False)
+
+        >>> pipe = KandinskyImg2ImgPipeline.from_pretrained(
+        ...     "kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16
+        ... )
+        >>> pipe.to("cuda")
+
+        >>> init_image = load_image(
+        ...     "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
+        ...     "/kandinsky/frog.png"
+        ... )
+
+        >>> image = pipe(
+        ...     prompt,
+        ...     image=init_image,
+        ...     image_embeds=image_emb,
+        ...     negative_image_embeds=zero_image_emb,
+        ...     height=768,
+        ...     width=768,
+        ...     num_inference_steps=100,
+        ...     strength=0.2,
+        ... ).images
+
+        >>> image[0].save("red_frog.png")
+        ```
+"""
+
+
+def get_new_h_w(h, w, scale_factor=8):
+    new_h = h // scale_factor**2
+    if h % scale_factor**2 != 0:
+        new_h += 1
+    new_w = w // scale_factor**2
+    if w % scale_factor**2 != 0:
+        new_w += 1
+    return new_h * scale_factor, new_w * scale_factor
+
+
+class KandinskyImg2ImgPipeline(DiffusionPipeline):
+    """
+    Pipeline for image-to-image generation using Kandinsky
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
+    library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
+
+    Args:
+        text_encoder ([`MultilingualCLIP`]):
+            Frozen text-encoder.
+        tokenizer ([`XLMRobertaTokenizer`]):
+            Tokenizer of class
+        scheduler ([`DDIMScheduler`]):
+            A scheduler to be used in combination with `unet` to generate image latents.
+        unet ([`UNet2DConditionModel`]):
+            Conditional U-Net architecture to denoise the image embedding.
+        movq ([`VQModel`]):
+            MoVQ image encoder and decoder
+    """
+
+    model_cpu_offload_seq = "text_encoder->unet->movq"
+
+    def __init__(
+        self,
+        text_encoder: MultilingualCLIP,
+        movq: VQModel,
+        tokenizer: XLMRobertaTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: DDIMScheduler,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            movq=movq,
+        )
+        self.movq_scale_factor = (
+            2 ** (len(self.movq.config.block_out_channels) - 1) if getattr(self, "movq", None) else 8
+        )
+        movq_latent_channels = self.movq.config.latent_channels if getattr(self, "movq", None) else 4
+        self.image_processor = VaeImageProcessor(
+            vae_scale_factor=self.movq_scale_factor,
+            vae_latent_channels=movq_latent_channels,
+            resample="bicubic",
+            reducing_gap=1,
+        )
+
+    def get_timesteps(self, num_inference_steps, strength, device):
+        # get the original timestep using init_timestep
+        init_timestep = min(int(num_inference_steps * strength), num_inference_steps)
+
+        t_start = max(num_inference_steps - init_timestep, 0)
+        timesteps = self.scheduler.timesteps[t_start:]
+
+        return timesteps, num_inference_steps - t_start
+
+    def prepare_latents(self, latents, latent_timestep, shape, dtype, device, generator, scheduler):
+        if latents is None:
+            latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        else:
+            if latents.shape != shape:
+                raise ValueError(f"Unexpected latents shape, got {latents.shape}, expected {shape}")
+            latents = latents.to(device)
+
+        latents = latents * scheduler.init_noise_sigma
+
+        shape = latents.shape
+        noise = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+
+        latents = self.add_noise(latents, noise, latent_timestep)
+        return latents
+
+    def _encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+    ):
+        batch_size = len(prompt) if isinstance(prompt, list) else 1
+        # get prompt text embeddings
+        text_inputs = self.tokenizer(
+            prompt,
+            padding="max_length",
+            max_length=77,
+            truncation=True,
+            return_attention_mask=True,
+            add_special_tokens=True,
+            return_tensors="pt",
+        )
+
+        text_input_ids = text_inputs.input_ids
+        untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+        if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(text_input_ids, untruncated_ids):
+            removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1])
+            logger.warning(
+                "The following part of your input was truncated because CLIP can only handle sequences up to"
+                f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+            )
+
+        text_input_ids = text_input_ids.to(device)
+        text_mask = text_inputs.attention_mask.to(device)
+
+        prompt_embeds, text_encoder_hidden_states = self.text_encoder(
+            input_ids=text_input_ids, attention_mask=text_mask
+        )
+
+        prompt_embeds = prompt_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+        text_encoder_hidden_states = text_encoder_hidden_states.repeat_interleave(num_images_per_prompt, dim=0)
+        text_mask = text_mask.repeat_interleave(num_images_per_prompt, dim=0)
+
+        if do_classifier_free_guidance:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=77,
+                truncation=True,
+                return_attention_mask=True,
+                add_special_tokens=True,
+                return_tensors="pt",
+            )
+            uncond_text_input_ids = uncond_input.input_ids.to(device)
+            uncond_text_mask = uncond_input.attention_mask.to(device)
+
+            negative_prompt_embeds, uncond_text_encoder_hidden_states = self.text_encoder(
+                input_ids=uncond_text_input_ids, attention_mask=uncond_text_mask
+            )
+
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+
+            seq_len = negative_prompt_embeds.shape[1]
+            negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt)
+            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len)
+
+            seq_len = uncond_text_encoder_hidden_states.shape[1]
+            uncond_text_encoder_hidden_states = uncond_text_encoder_hidden_states.repeat(1, num_images_per_prompt, 1)
+            uncond_text_encoder_hidden_states = uncond_text_encoder_hidden_states.view(
+                batch_size * num_images_per_prompt, seq_len, -1
+            )
+            uncond_text_mask = uncond_text_mask.repeat_interleave(num_images_per_prompt, dim=0)
+
+            # done duplicates
+
+            # For classifier free guidance, we need to do two forward passes.
+            # Here we concatenate the unconditional and text embeddings into a single batch
+            # to avoid doing two forward passes
+            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
+            text_encoder_hidden_states = torch.cat([uncond_text_encoder_hidden_states, text_encoder_hidden_states])
+
+            text_mask = torch.cat([uncond_text_mask, text_mask])
+
+        return prompt_embeds, text_encoder_hidden_states, text_mask
+
+    #  add_noise method to overwrite the one in schedule because it use a different beta schedule for adding noise vs sampling
+    def add_noise(
+        self,
+        original_samples: torch.Tensor,
+        noise: torch.Tensor,
+        timesteps: torch.IntTensor,
+    ) -> torch.Tensor:
+        betas = torch.linspace(0.0001, 0.02, 1000, dtype=torch.float32)
+        alphas = 1.0 - betas
+        alphas_cumprod = torch.cumprod(alphas, dim=0)
+        alphas_cumprod = alphas_cumprod.to(device=original_samples.device, dtype=original_samples.dtype)
+        timesteps = timesteps.to(original_samples.device)
+
+        sqrt_alpha_prod = alphas_cumprod[timesteps] ** 0.5
+        sqrt_alpha_prod = sqrt_alpha_prod.flatten()
+        while len(sqrt_alpha_prod.shape) < len(original_samples.shape):
+            sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze(-1)
+
+        sqrt_one_minus_alpha_prod = (1 - alphas_cumprod[timesteps]) ** 0.5
+        sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten()
+        while len(sqrt_one_minus_alpha_prod.shape) < len(original_samples.shape):
+            sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze(-1)
+
+        noisy_samples = sqrt_alpha_prod * original_samples + sqrt_one_minus_alpha_prod * noise
+
+        return noisy_samples
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
+        image_embeds: torch.Tensor,
+        negative_image_embeds: torch.Tensor,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        height: int = 512,
+        width: int = 512,
+        num_inference_steps: int = 100,
+        strength: float = 0.3,
+        guidance_scale: float = 7.0,
+        num_images_per_prompt: int = 1,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        output_type: Optional[str] = "pil",
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
+        callback_steps: int = 1,
+        return_dict: bool = True,
+    ):
+        """
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`):
+                The prompt or prompts to guide the image generation.
+            image (`torch.Tensor`, `PIL.Image.Image`):
+                `Image`, or tensor representing an image batch, that will be used as the starting point for the
+                process.
+            image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
+                The clip image embeddings for text prompt, that will be used to condition the image generation.
+            negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
+                The clip image embeddings for negative text prompt, will be used to condition the image generation.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
+                if `guidance_scale` is less than `1`).
+            height (`int`, *optional*, defaults to 512):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to 512):
+                The width in pixels of the generated image.
+            num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            strength (`float`, *optional*, defaults to 0.3):
+                Conceptually, indicates how much to transform the reference `image`. Must be between 0 and 1. `image`
+                will be used as a starting point, adding more noise to it the larger the `strength`. The number of
+                denoising steps depends on the amount of noise initially added. When `strength` is 1, added noise will
+                be maximum and the denoising process will run for the full number of iterations specified in
+                `num_inference_steps`. A value of 1, therefore, essentially ignores `image`.
+            guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            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.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generate image. Choose between: `"pil"` (`PIL.Image.Image`), `"np"`
+                (`np.array`) or `"pt"` (`torch.Tensor`).
+            callback (`Callable`, *optional*):
+                A function that calls every `callback_steps` steps during inference. The function is 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 is called. If not specified, the callback is called at
+                every step.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
+
+        Examples:
+
+        Returns:
+            [`~pipelines.ImagePipelineOutput`] or `tuple`
+        """
+        # 1. Define call parameters
+        if isinstance(prompt, str):
+            batch_size = 1
+        elif isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        device = self._execution_device
+
+        batch_size = batch_size * num_images_per_prompt
+
+        do_classifier_free_guidance = guidance_scale > 1.0
+
+        # 2. get text and image embeddings
+        prompt_embeds, text_encoder_hidden_states, _ = self._encode_prompt(
+            prompt, device, num_images_per_prompt, do_classifier_free_guidance, negative_prompt
+        )
+
+        if isinstance(image_embeds, list):
+            image_embeds = torch.cat(image_embeds, dim=0)
+        if isinstance(negative_image_embeds, list):
+            negative_image_embeds = torch.cat(negative_image_embeds, dim=0)
+
+        if do_classifier_free_guidance:
+            image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+            negative_image_embeds = negative_image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+
+            image_embeds = torch.cat([negative_image_embeds, image_embeds], dim=0).to(
+                dtype=prompt_embeds.dtype, device=device
+            )
+
+        # 3. pre-processing initial image
+        if not isinstance(image, list):
+            image = [image]
+        if not all(isinstance(i, (PIL.Image.Image, torch.Tensor)) for i in image):
+            raise ValueError(
+                f"Input is in incorrect format: {[type(i) for i in image]}. Currently, we only support  PIL image and pytorch tensor"
+            )
+
+        image = torch.cat([self.image_processor.preprocess(i, width, height) for i in image], dim=0)
+        image = image.to(dtype=prompt_embeds.dtype, device=device)
+
+        latents = self.movq.encode(image)["latents"]
+        latents = latents.repeat_interleave(num_images_per_prompt, dim=0)
+
+        # 4. set timesteps
+        self.scheduler.set_timesteps(num_inference_steps, device=device)
+
+        timesteps_tensor, num_inference_steps = self.get_timesteps(num_inference_steps, strength, device)
+
+        # the formular to calculate timestep for add_noise is taken from the original kandinsky repo
+        latent_timestep = int(self.scheduler.config.num_train_timesteps * strength) - 2
+
+        latent_timestep = torch.tensor([latent_timestep] * batch_size, dtype=timesteps_tensor.dtype, device=device)
+
+        num_channels_latents = self.unet.config.in_channels
+
+        height, width = get_new_h_w(height, width, self.movq_scale_factor)
+
+        # 5. Create initial latent
+        latents = self.prepare_latents(
+            latents,
+            latent_timestep,
+            (batch_size, num_channels_latents, height, width),
+            text_encoder_hidden_states.dtype,
+            device,
+            generator,
+            self.scheduler,
+        )
+
+        # 6. Denoising loop
+        for i, t in enumerate(self.progress_bar(timesteps_tensor)):
+            # expand the latents if we are doing classifier free guidance
+            latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
+
+            added_cond_kwargs = {"text_embeds": prompt_embeds, "image_embeds": image_embeds}
+            noise_pred = self.unet(
+                sample=latent_model_input,
+                timestep=t,
+                encoder_hidden_states=text_encoder_hidden_states,
+                added_cond_kwargs=added_cond_kwargs,
+                return_dict=False,
+            )[0]
+
+            if do_classifier_free_guidance:
+                noise_pred, variance_pred = noise_pred.split(latents.shape[1], dim=1)
+                noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
+                _, variance_pred_text = variance_pred.chunk(2)
+                noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
+                noise_pred = torch.cat([noise_pred, variance_pred_text], dim=1)
+
+            if not (
+                hasattr(self.scheduler.config, "variance_type")
+                and self.scheduler.config.variance_type in ["learned", "learned_range"]
+            ):
+                noise_pred, _ = noise_pred.split(latents.shape[1], dim=1)
+
+            # compute the previous noisy sample x_t -> x_t-1
+            latents = self.scheduler.step(
+                noise_pred,
+                t,
+                latents,
+                generator=generator,
+            ).prev_sample
+
+            if callback is not None and i % callback_steps == 0:
+                step_idx = i // getattr(self.scheduler, "order", 1)
+                callback(step_idx, t, latents)
+
+            if XLA_AVAILABLE:
+                xm.mark_step()
+
+        # 7. post-processing
+        image = self.movq.decode(latents, force_not_quantize=True)["sample"]
+
+        self.maybe_free_model_hooks()
+
+        if output_type not in ["pt", "np", "pil"]:
+            raise ValueError(f"Only the output types `pt`, `pil` and `np` are supported not output_type={output_type}")
+
+        image = self.image_processor.postprocess(image, output_type)
+
+        if not return_dict:
+            return (image,)
+
+        return ImagePipelineOutput(images=image)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/kandinsky/pipeline_kandinsky_inpaint.py

@@ -0,0 +1,647 @@
+# Copyright 2025 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.
+
+from copy import deepcopy
+from typing import Callable, List, Optional, Union
+
+import numpy as np
+import PIL.Image
+import torch
+import torch.nn.functional as F
+from packaging import version
+from PIL import Image
+from transformers import (
+    XLMRobertaTokenizer,
+)
+
+from ... import __version__
+from ...models import UNet2DConditionModel, VQModel
+from ...schedulers import DDIMScheduler
+from ...utils import (
+    is_torch_xla_available,
+    logging,
+    replace_example_docstring,
+)
+from ...utils.torch_utils import randn_tensor
+from ..pipeline_utils import DiffusionPipeline, ImagePipelineOutput
+from .text_encoder import MultilingualCLIP
+
+
+if is_torch_xla_available():
+    import torch_xla.core.xla_model as xm
+
+    XLA_AVAILABLE = True
+else:
+    XLA_AVAILABLE = False
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+
+EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> from diffusers import KandinskyInpaintPipeline, KandinskyPriorPipeline
+        >>> from diffusers.utils import load_image
+        >>> import torch
+        >>> import numpy as np
+
+        >>> pipe_prior = KandinskyPriorPipeline.from_pretrained(
+        ...     "kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16
+        ... )
+        >>> pipe_prior.to("cuda")
+
+        >>> prompt = "a hat"
+        >>> image_emb, zero_image_emb = pipe_prior(prompt, return_dict=False)
+
+        >>> pipe = KandinskyInpaintPipeline.from_pretrained(
+        ...     "kandinsky-community/kandinsky-2-1-inpaint", torch_dtype=torch.float16
+        ... )
+        >>> pipe.to("cuda")
+
+        >>> init_image = load_image(
+        ...     "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
+        ...     "/kandinsky/cat.png"
+        ... )
+
+        >>> mask = np.zeros((768, 768), dtype=np.float32)
+        >>> mask[:250, 250:-250] = 1
+
+        >>> out = pipe(
+        ...     prompt,
+        ...     image=init_image,
+        ...     mask_image=mask,
+        ...     image_embeds=image_emb,
+        ...     negative_image_embeds=zero_image_emb,
+        ...     height=768,
+        ...     width=768,
+        ...     num_inference_steps=50,
+        ... )
+
+        >>> image = out.images[0]
+        >>> image.save("cat_with_hat.png")
+        ```
+"""
+
+
+def get_new_h_w(h, w, scale_factor=8):
+    new_h = h // scale_factor**2
+    if h % scale_factor**2 != 0:
+        new_h += 1
+    new_w = w // scale_factor**2
+    if w % scale_factor**2 != 0:
+        new_w += 1
+    return new_h * scale_factor, new_w * scale_factor
+
+
+def prepare_mask(masks):
+    prepared_masks = []
+    for mask in masks:
+        old_mask = deepcopy(mask)
+        for i in range(mask.shape[1]):
+            for j in range(mask.shape[2]):
+                if old_mask[0][i][j] == 1:
+                    continue
+                if i != 0:
+                    mask[:, i - 1, j] = 0
+                if j != 0:
+                    mask[:, i, j - 1] = 0
+                if i != 0 and j != 0:
+                    mask[:, i - 1, j - 1] = 0
+                if i != mask.shape[1] - 1:
+                    mask[:, i + 1, j] = 0
+                if j != mask.shape[2] - 1:
+                    mask[:, i, j + 1] = 0
+                if i != mask.shape[1] - 1 and j != mask.shape[2] - 1:
+                    mask[:, i + 1, j + 1] = 0
+        prepared_masks.append(mask)
+    return torch.stack(prepared_masks, dim=0)
+
+
+def prepare_mask_and_masked_image(image, mask, height, width):
+    r"""
+    Prepares a pair (mask, image) to be consumed by the Kandinsky inpaint pipeline. This means that those inputs will
+    be converted to ``torch.Tensor`` with shapes ``batch x channels x height x width`` where ``channels`` is ``3`` for
+    the ``image`` and ``1`` for the ``mask``.
+
+    The ``image`` will be converted to ``torch.float32`` and normalized to be in ``[-1, 1]``. The ``mask`` will be
+    binarized (``mask > 0.5``) and cast to ``torch.float32`` too.
+
+    Args:
+        image (Union[np.array, PIL.Image, torch.Tensor]): The image to inpaint.
+            It can be a ``PIL.Image``, or a ``height x width x 3`` ``np.array`` or a ``channels x height x width``
+            ``torch.Tensor`` or a ``batch x channels x height x width`` ``torch.Tensor``.
+        mask (_type_): The mask to apply to the image, i.e. regions to inpaint.
+            It can be a ``PIL.Image``, or a ``height x width`` ``np.array`` or a ``1 x height x width``
+            ``torch.Tensor`` or a ``batch x 1 x height x width`` ``torch.Tensor``.
+        height (`int`, *optional*, defaults to 512):
+            The height in pixels of the generated image.
+        width (`int`, *optional*, defaults to 512):
+            The width in pixels of the generated image.
+
+
+    Raises:
+        ValueError: ``torch.Tensor`` images should be in the ``[-1, 1]`` range. ValueError: ``torch.Tensor`` mask
+        should be in the ``[0, 1]`` range. ValueError: ``mask`` and ``image`` should have the same spatial dimensions.
+        TypeError: ``mask`` is a ``torch.Tensor`` but ``image`` is not
+            (ot the other way around).
+
+    Returns:
+        tuple[torch.Tensor]: The pair (mask, image) as ``torch.Tensor`` with 4
+            dimensions: ``batch x channels x height x width``.
+    """
+
+    if image is None:
+        raise ValueError("`image` input cannot be undefined.")
+
+    if mask is None:
+        raise ValueError("`mask_image` input cannot be undefined.")
+
+    if isinstance(image, torch.Tensor):
+        if not isinstance(mask, torch.Tensor):
+            raise TypeError(f"`image` is a torch.Tensor but `mask` (type: {type(mask)} is not")
+
+        # Batch single image
+        if image.ndim == 3:
+            assert image.shape[0] == 3, "Image outside a batch should be of shape (3, H, W)"
+            image = image.unsqueeze(0)
+
+        # Batch and add channel dim for single mask
+        if mask.ndim == 2:
+            mask = mask.unsqueeze(0).unsqueeze(0)
+
+        # Batch single mask or add channel dim
+        if mask.ndim == 3:
+            # Single batched mask, no channel dim or single mask not batched but channel dim
+            if mask.shape[0] == 1:
+                mask = mask.unsqueeze(0)
+
+            # Batched masks no channel dim
+            else:
+                mask = mask.unsqueeze(1)
+
+        assert image.ndim == 4 and mask.ndim == 4, "Image and Mask must have 4 dimensions"
+        assert image.shape[-2:] == mask.shape[-2:], "Image and Mask must have the same spatial dimensions"
+        assert image.shape[0] == mask.shape[0], "Image and Mask must have the same batch size"
+
+        # Check image is in [-1, 1]
+        if image.min() < -1 or image.max() > 1:
+            raise ValueError("Image should be in [-1, 1] range")
+
+        # Check mask is in [0, 1]
+        if mask.min() < 0 or mask.max() > 1:
+            raise ValueError("Mask should be in [0, 1] range")
+
+        # Binarize mask
+        mask[mask < 0.5] = 0
+        mask[mask >= 0.5] = 1
+
+        # Image as float32
+        image = image.to(dtype=torch.float32)
+    elif isinstance(mask, torch.Tensor):
+        raise TypeError(f"`mask` is a torch.Tensor but `image` (type: {type(image)} is not")
+    else:
+        # preprocess image
+        if isinstance(image, (PIL.Image.Image, np.ndarray)):
+            image = [image]
+
+        if isinstance(image, list) and isinstance(image[0], PIL.Image.Image):
+            # resize all images w.r.t passed height an width
+            image = [i.resize((width, height), resample=Image.BICUBIC, reducing_gap=1) for i in image]
+            image = [np.array(i.convert("RGB"))[None, :] for i in image]
+            image = np.concatenate(image, axis=0)
+        elif isinstance(image, list) and isinstance(image[0], np.ndarray):
+            image = np.concatenate([i[None, :] for i in image], axis=0)
+
+        image = image.transpose(0, 3, 1, 2)
+        image = torch.from_numpy(image).to(dtype=torch.float32) / 127.5 - 1.0
+
+        # preprocess mask
+        if isinstance(mask, (PIL.Image.Image, np.ndarray)):
+            mask = [mask]
+
+        if isinstance(mask, list) and isinstance(mask[0], PIL.Image.Image):
+            mask = [i.resize((width, height), resample=PIL.Image.LANCZOS) for i in mask]
+            mask = np.concatenate([np.array(m.convert("L"))[None, None, :] for m in mask], axis=0)
+            mask = mask.astype(np.float32) / 255.0
+        elif isinstance(mask, list) and isinstance(mask[0], np.ndarray):
+            mask = np.concatenate([m[None, None, :] for m in mask], axis=0)
+
+        mask[mask < 0.5] = 0
+        mask[mask >= 0.5] = 1
+        mask = torch.from_numpy(mask)
+
+    mask = 1 - mask
+
+    return mask, image
+
+
+class KandinskyInpaintPipeline(DiffusionPipeline):
+    """
+    Pipeline for text-guided image inpainting using Kandinsky2.1
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
+    library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
+
+    Args:
+        text_encoder ([`MultilingualCLIP`]):
+            Frozen text-encoder.
+        tokenizer ([`XLMRobertaTokenizer`]):
+            Tokenizer of class
+        scheduler ([`DDIMScheduler`]):
+            A scheduler to be used in combination with `unet` to generate image latents.
+        unet ([`UNet2DConditionModel`]):
+            Conditional U-Net architecture to denoise the image embedding.
+        movq ([`VQModel`]):
+            MoVQ image encoder and decoder
+    """
+
+    model_cpu_offload_seq = "text_encoder->unet->movq"
+
+    def __init__(
+        self,
+        text_encoder: MultilingualCLIP,
+        movq: VQModel,
+        tokenizer: XLMRobertaTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: DDIMScheduler,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            text_encoder=text_encoder,
+            movq=movq,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+        )
+        self.movq_scale_factor = 2 ** (len(self.movq.config.block_out_channels) - 1)
+        self._warn_has_been_called = False
+
+    # Copied from diffusers.pipelines.unclip.pipeline_unclip.UnCLIPPipeline.prepare_latents
+    def prepare_latents(self, shape, dtype, device, generator, latents, scheduler):
+        if latents is None:
+            latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        else:
+            if latents.shape != shape:
+                raise ValueError(f"Unexpected latents shape, got {latents.shape}, expected {shape}")
+            latents = latents.to(device)
+
+        latents = latents * scheduler.init_noise_sigma
+        return latents
+
+    def _encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+    ):
+        batch_size = len(prompt) if isinstance(prompt, list) else 1
+        # get prompt text embeddings
+        text_inputs = self.tokenizer(
+            prompt,
+            padding="max_length",
+            max_length=77,
+            truncation=True,
+            return_attention_mask=True,
+            add_special_tokens=True,
+            return_tensors="pt",
+        )
+
+        text_input_ids = text_inputs.input_ids
+        untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+        if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(text_input_ids, untruncated_ids):
+            removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1])
+            logger.warning(
+                "The following part of your input was truncated because CLIP can only handle sequences up to"
+                f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+            )
+
+        text_input_ids = text_input_ids.to(device)
+        text_mask = text_inputs.attention_mask.to(device)
+
+        prompt_embeds, text_encoder_hidden_states = self.text_encoder(
+            input_ids=text_input_ids, attention_mask=text_mask
+        )
+
+        prompt_embeds = prompt_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+        text_encoder_hidden_states = text_encoder_hidden_states.repeat_interleave(num_images_per_prompt, dim=0)
+        text_mask = text_mask.repeat_interleave(num_images_per_prompt, dim=0)
+
+        if do_classifier_free_guidance:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=77,
+                truncation=True,
+                return_attention_mask=True,
+                add_special_tokens=True,
+                return_tensors="pt",
+            )
+            uncond_text_input_ids = uncond_input.input_ids.to(device)
+            uncond_text_mask = uncond_input.attention_mask.to(device)
+
+            negative_prompt_embeds, uncond_text_encoder_hidden_states = self.text_encoder(
+                input_ids=uncond_text_input_ids, attention_mask=uncond_text_mask
+            )
+
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+
+            seq_len = negative_prompt_embeds.shape[1]
+            negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt)
+            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len)
+
+            seq_len = uncond_text_encoder_hidden_states.shape[1]
+            uncond_text_encoder_hidden_states = uncond_text_encoder_hidden_states.repeat(1, num_images_per_prompt, 1)
+            uncond_text_encoder_hidden_states = uncond_text_encoder_hidden_states.view(
+                batch_size * num_images_per_prompt, seq_len, -1
+            )
+            uncond_text_mask = uncond_text_mask.repeat_interleave(num_images_per_prompt, dim=0)
+
+            # done duplicates
+
+            # For classifier free guidance, we need to do two forward passes.
+            # Here we concatenate the unconditional and text embeddings into a single batch
+            # to avoid doing two forward passes
+            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
+            text_encoder_hidden_states = torch.cat([uncond_text_encoder_hidden_states, text_encoder_hidden_states])
+
+            text_mask = torch.cat([uncond_text_mask, text_mask])
+
+        return prompt_embeds, text_encoder_hidden_states, text_mask
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        image: Union[torch.Tensor, PIL.Image.Image],
+        mask_image: Union[torch.Tensor, PIL.Image.Image, np.ndarray],
+        image_embeds: torch.Tensor,
+        negative_image_embeds: torch.Tensor,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        height: int = 512,
+        width: int = 512,
+        num_inference_steps: int = 100,
+        guidance_scale: float = 4.0,
+        num_images_per_prompt: int = 1,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: Optional[str] = "pil",
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
+        callback_steps: int = 1,
+        return_dict: bool = True,
+    ):
+        """
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`):
+                The prompt or prompts to guide the image generation.
+            image (`torch.Tensor`, `PIL.Image.Image` or `np.ndarray`):
+                `Image`, or tensor representing an image batch, that will be used as the starting point for the
+                process.
+            mask_image (`PIL.Image.Image`,`torch.Tensor` or `np.ndarray`):
+                `Image`, or a tensor representing an image batch, to mask `image`. White pixels in the mask will be
+                repainted, while black pixels will be preserved. You can pass a pytorch tensor as mask only if the
+                image you passed is a pytorch tensor, and it should contain one color channel (L) instead of 3, so the
+                expected shape would be either `(B, 1, H, W,)`, `(B, H, W)`, `(1, H, W)` or `(H, W)` If image is an PIL
+                image or numpy array, mask should also be a either PIL image or numpy array. If it is a PIL image, it
+                will be converted to a single channel (luminance) before use. If it is a nummpy array, the expected
+                shape is `(H, W)`.
+            image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
+                The clip image embeddings for text prompt, that will be used to condition the image generation.
+            negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
+                The clip image embeddings for negative text prompt, will be used to condition the image generation.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
+                if `guidance_scale` is less than `1`).
+            height (`int`, *optional*, defaults to 512):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to 512):
+                The width in pixels of the generated image.
+            num_inference_steps (`int`, *optional*, defaults to 100):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            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.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 be generated by sampling using the supplied random `generator`.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generate image. Choose between: `"pil"` (`PIL.Image.Image`), `"np"`
+                (`np.array`) or `"pt"` (`torch.Tensor`).
+            callback (`Callable`, *optional*):
+                A function that calls every `callback_steps` steps during inference. The function is 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 is called. If not specified, the callback is called at
+                every step.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
+
+        Examples:
+
+        Returns:
+            [`~pipelines.ImagePipelineOutput`] or `tuple`
+        """
+        if not self._warn_has_been_called and version.parse(version.parse(__version__).base_version) < version.parse(
+            "0.23.0.dev0"
+        ):
+            logger.warning(
+                "Please note that the expected format of `mask_image` has recently been changed. "
+                "Before diffusers == 0.19.0, Kandinsky Inpainting pipelines repainted black pixels and preserved black pixels. "
+                "As of diffusers==0.19.0 this behavior has been inverted. Now white pixels are repainted and black pixels are preserved. "
+                "This way, Kandinsky's masking behavior is aligned with Stable Diffusion. "
+                "THIS means that you HAVE to invert the input mask to have the same behavior as before as explained in https://github.com/huggingface/diffusers/pull/4207. "
+                "This warning will be suppressed after the first inference call and will be removed in diffusers>0.23.0"
+            )
+            self._warn_has_been_called = True
+
+        # Define call parameters
+        if isinstance(prompt, str):
+            batch_size = 1
+        elif isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        device = self._execution_device
+
+        batch_size = batch_size * num_images_per_prompt
+        do_classifier_free_guidance = guidance_scale > 1.0
+
+        prompt_embeds, text_encoder_hidden_states, _ = self._encode_prompt(
+            prompt, device, num_images_per_prompt, do_classifier_free_guidance, negative_prompt
+        )
+
+        if isinstance(image_embeds, list):
+            image_embeds = torch.cat(image_embeds, dim=0)
+        if isinstance(negative_image_embeds, list):
+            negative_image_embeds = torch.cat(negative_image_embeds, dim=0)
+
+        if do_classifier_free_guidance:
+            image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+            negative_image_embeds = negative_image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+
+            image_embeds = torch.cat([negative_image_embeds, image_embeds], dim=0).to(
+                dtype=prompt_embeds.dtype, device=device
+            )
+
+        # preprocess image and mask
+        mask_image, image = prepare_mask_and_masked_image(image, mask_image, height, width)
+
+        image = image.to(dtype=prompt_embeds.dtype, device=device)
+        image = self.movq.encode(image)["latents"]
+
+        mask_image = mask_image.to(dtype=prompt_embeds.dtype, device=device)
+
+        image_shape = tuple(image.shape[-2:])
+        mask_image = F.interpolate(
+            mask_image,
+            image_shape,
+            mode="nearest",
+        )
+        mask_image = prepare_mask(mask_image)
+        masked_image = image * mask_image
+
+        mask_image = mask_image.repeat_interleave(num_images_per_prompt, dim=0)
+        masked_image = masked_image.repeat_interleave(num_images_per_prompt, dim=0)
+        if do_classifier_free_guidance:
+            mask_image = mask_image.repeat(2, 1, 1, 1)
+            masked_image = masked_image.repeat(2, 1, 1, 1)
+
+        self.scheduler.set_timesteps(num_inference_steps, device=device)
+        timesteps_tensor = self.scheduler.timesteps
+
+        num_channels_latents = self.movq.config.latent_channels
+
+        # get h, w for latents
+        sample_height, sample_width = get_new_h_w(height, width, self.movq_scale_factor)
+
+        # create initial latent
+        latents = self.prepare_latents(
+            (batch_size, num_channels_latents, sample_height, sample_width),
+            text_encoder_hidden_states.dtype,
+            device,
+            generator,
+            latents,
+            self.scheduler,
+        )
+
+        # Check that sizes of mask, masked image and latents match with expected
+        num_channels_mask = mask_image.shape[1]
+        num_channels_masked_image = masked_image.shape[1]
+        if num_channels_latents + num_channels_mask + num_channels_masked_image != self.unet.config.in_channels:
+            raise ValueError(
+                f"Incorrect configuration settings! The config of `pipeline.unet`: {self.unet.config} expects"
+                f" {self.unet.config.in_channels} but received `num_channels_latents`: {num_channels_latents} +"
+                f" `num_channels_mask`: {num_channels_mask} + `num_channels_masked_image`: {num_channels_masked_image}"
+                f" = {num_channels_latents + num_channels_masked_image + num_channels_mask}. Please verify the config of"
+                " `pipeline.unet` or your `mask_image` or `image` input."
+            )
+
+        for i, t in enumerate(self.progress_bar(timesteps_tensor)):
+            # expand the latents if we are doing classifier free guidance
+            latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
+            latent_model_input = torch.cat([latent_model_input, masked_image, mask_image], dim=1)
+
+            added_cond_kwargs = {"text_embeds": prompt_embeds, "image_embeds": image_embeds}
+            noise_pred = self.unet(
+                sample=latent_model_input,
+                timestep=t,
+                encoder_hidden_states=text_encoder_hidden_states,
+                added_cond_kwargs=added_cond_kwargs,
+                return_dict=False,
+            )[0]
+
+            if do_classifier_free_guidance:
+                noise_pred, variance_pred = noise_pred.split(latents.shape[1], dim=1)
+                noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
+                _, variance_pred_text = variance_pred.chunk(2)
+                noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
+                noise_pred = torch.cat([noise_pred, variance_pred_text], dim=1)
+
+            if not (
+                hasattr(self.scheduler.config, "variance_type")
+                and self.scheduler.config.variance_type in ["learned", "learned_range"]
+            ):
+                noise_pred, _ = noise_pred.split(latents.shape[1], dim=1)
+
+            # compute the previous noisy sample x_t -> x_t-1
+            latents = self.scheduler.step(
+                noise_pred,
+                t,
+                latents,
+                generator=generator,
+            ).prev_sample
+
+            if callback is not None and i % callback_steps == 0:
+                step_idx = i // getattr(self.scheduler, "order", 1)
+                callback(step_idx, t, latents)
+
+            if XLA_AVAILABLE:
+                xm.mark_step()
+
+        # post-processing
+        image = self.movq.decode(latents, force_not_quantize=True)["sample"]
+
+        self.maybe_free_model_hooks()
+
+        if output_type not in ["pt", "np", "pil"]:
+            raise ValueError(f"Only the output types `pt`, `pil` and `np` are supported not output_type={output_type}")
+
+        if output_type in ["np", "pil"]:
+            image = image * 0.5 + 0.5
+            image = image.clamp(0, 1)
+            image = image.cpu().permute(0, 2, 3, 1).float().numpy()
+
+        if output_type == "pil":
+            image = self.numpy_to_pil(image)
+
+        if not return_dict:
+            return (image,)
+
+        return ImagePipelineOutput(images=image)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/kandinsky/pipeline_kandinsky_prior.py

@@ -0,0 +1,559 @@
+# Copyright 2025 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.
+
+from dataclasses import dataclass
+from typing import List, Optional, Union
+
+import numpy as np
+import PIL.Image
+import torch
+from transformers import CLIPImageProcessor, CLIPTextModelWithProjection, CLIPTokenizer, CLIPVisionModelWithProjection
+
+from ...models import PriorTransformer
+from ...schedulers import UnCLIPScheduler
+from ...utils import (
+    BaseOutput,
+    is_torch_xla_available,
+    logging,
+    replace_example_docstring,
+)
+from ...utils.torch_utils import randn_tensor
+from ..pipeline_utils import DiffusionPipeline
+
+
+if is_torch_xla_available():
+    import torch_xla.core.xla_model as xm
+
+    XLA_AVAILABLE = True
+else:
+    XLA_AVAILABLE = False
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+
+EXAMPLE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> from diffusers import KandinskyPipeline, KandinskyPriorPipeline
+        >>> import torch
+
+        >>> pipe_prior = KandinskyPriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-1-prior")
+        >>> pipe_prior.to("cuda")
+
+        >>> prompt = "red cat, 4k photo"
+        >>> out = pipe_prior(prompt)
+        >>> image_emb = out.image_embeds
+        >>> negative_image_emb = out.negative_image_embeds
+
+        >>> pipe = KandinskyPipeline.from_pretrained("kandinsky-community/kandinsky-2-1")
+        >>> pipe.to("cuda")
+
+        >>> image = pipe(
+        ...     prompt,
+        ...     image_embeds=image_emb,
+        ...     negative_image_embeds=negative_image_emb,
+        ...     height=768,
+        ...     width=768,
+        ...     num_inference_steps=100,
+        ... ).images
+
+        >>> image[0].save("cat.png")
+        ```
+"""
+
+EXAMPLE_INTERPOLATE_DOC_STRING = """
+    Examples:
+        ```py
+        >>> from diffusers import KandinskyPriorPipeline, KandinskyPipeline
+        >>> from diffusers.utils import load_image
+        >>> import PIL
+
+        >>> import torch
+        >>> from torchvision import transforms
+
+        >>> pipe_prior = KandinskyPriorPipeline.from_pretrained(
+        ...     "kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16
+        ... )
+        >>> pipe_prior.to("cuda")
+
+        >>> img1 = load_image(
+        ...     "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
+        ...     "/kandinsky/cat.png"
+        ... )
+
+        >>> img2 = load_image(
+        ...     "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
+        ...     "/kandinsky/starry_night.jpeg"
+        ... )
+
+        >>> images_texts = ["a cat", img1, img2]
+        >>> weights = [0.3, 0.3, 0.4]
+        >>> image_emb, zero_image_emb = pipe_prior.interpolate(images_texts, weights)
+
+        >>> pipe = KandinskyPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
+        >>> pipe.to("cuda")
+
+        >>> image = pipe(
+        ...     "",
+        ...     image_embeds=image_emb,
+        ...     negative_image_embeds=zero_image_emb,
+        ...     height=768,
+        ...     width=768,
+        ...     num_inference_steps=150,
+        ... ).images[0]
+
+        >>> image.save("starry_cat.png")
+        ```
+"""
+
+
+@dataclass
+class KandinskyPriorPipelineOutput(BaseOutput):
+    """
+    Output class for KandinskyPriorPipeline.
+
+    Args:
+        image_embeds (`torch.Tensor`)
+            clip image embeddings for text prompt
+        negative_image_embeds (`List[PIL.Image.Image]` or `np.ndarray`)
+            clip image embeddings for unconditional tokens
+    """
+
+    image_embeds: Union[torch.Tensor, np.ndarray]
+    negative_image_embeds: Union[torch.Tensor, np.ndarray]
+
+
+class KandinskyPriorPipeline(DiffusionPipeline):
+    """
+    Pipeline for generating image prior for Kandinsky
+
+    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
+    library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
+
+    Args:
+        prior ([`PriorTransformer`]):
+            The canonical unCLIP prior to approximate the image embedding from the text embedding.
+        image_encoder ([`CLIPVisionModelWithProjection`]):
+            Frozen image-encoder.
+        text_encoder ([`CLIPTextModelWithProjection`]):
+            Frozen text-encoder.
+        tokenizer (`CLIPTokenizer`):
+            Tokenizer of class
+            [CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer).
+        scheduler ([`UnCLIPScheduler`]):
+            A scheduler to be used in combination with `prior` to generate image embedding.
+    """
+
+    _exclude_from_cpu_offload = ["prior"]
+    model_cpu_offload_seq = "text_encoder->prior"
+
+    def __init__(
+        self,
+        prior: PriorTransformer,
+        image_encoder: CLIPVisionModelWithProjection,
+        text_encoder: CLIPTextModelWithProjection,
+        tokenizer: CLIPTokenizer,
+        scheduler: UnCLIPScheduler,
+        image_processor: CLIPImageProcessor,
+    ):
+        super().__init__()
+
+        self.register_modules(
+            prior=prior,
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            scheduler=scheduler,
+            image_encoder=image_encoder,
+            image_processor=image_processor,
+        )
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_INTERPOLATE_DOC_STRING)
+    def interpolate(
+        self,
+        images_and_prompts: List[Union[str, PIL.Image.Image, torch.Tensor]],
+        weights: List[float],
+        num_images_per_prompt: int = 1,
+        num_inference_steps: int = 25,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        negative_prior_prompt: Optional[str] = None,
+        negative_prompt: str = "",
+        guidance_scale: float = 4.0,
+        device=None,
+    ):
+        """
+        Function invoked when using the prior pipeline for interpolation.
+
+        Args:
+            images_and_prompts (`List[Union[str, PIL.Image.Image, torch.Tensor]]`):
+                list of prompts and images to guide the image generation.
+            weights: (`List[float]`):
+                list of weights for each condition in `images_and_prompts`
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            num_inference_steps (`int`, *optional*, defaults to 25):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            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.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 be generated by sampling using the supplied random `generator`.
+            negative_prior_prompt (`str`, *optional*):
+                The prompt not to guide the prior diffusion process. Ignored when not using guidance (i.e., ignored if
+                `guidance_scale` is less than `1`).
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt not to guide the image generation. Ignored when not using guidance (i.e., ignored if
+                `guidance_scale` is less than `1`).
+            guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+
+        Examples:
+
+        Returns:
+            [`KandinskyPriorPipelineOutput`] or `tuple`
+        """
+
+        device = device or self.device
+
+        if len(images_and_prompts) != len(weights):
+            raise ValueError(
+                f"`images_and_prompts` contains {len(images_and_prompts)} items and `weights` contains {len(weights)} items - they should be lists of same length"
+            )
+
+        image_embeddings = []
+        for cond, weight in zip(images_and_prompts, weights):
+            if isinstance(cond, str):
+                image_emb = self(
+                    cond,
+                    num_inference_steps=num_inference_steps,
+                    num_images_per_prompt=num_images_per_prompt,
+                    generator=generator,
+                    latents=latents,
+                    negative_prompt=negative_prior_prompt,
+                    guidance_scale=guidance_scale,
+                ).image_embeds
+
+            elif isinstance(cond, (PIL.Image.Image, torch.Tensor)):
+                if isinstance(cond, PIL.Image.Image):
+                    cond = (
+                        self.image_processor(cond, return_tensors="pt")
+                        .pixel_values[0]
+                        .unsqueeze(0)
+                        .to(dtype=self.image_encoder.dtype, device=device)
+                    )
+
+                image_emb = self.image_encoder(cond)["image_embeds"]
+
+            else:
+                raise ValueError(
+                    f"`images_and_prompts` can only contains elements to be of type `str`, `PIL.Image.Image` or `torch.Tensor`  but is {type(cond)}"
+                )
+
+            image_embeddings.append(image_emb * weight)
+
+        image_emb = torch.cat(image_embeddings).sum(dim=0, keepdim=True)
+
+        out_zero = self(
+            negative_prompt,
+            num_inference_steps=num_inference_steps,
+            num_images_per_prompt=num_images_per_prompt,
+            generator=generator,
+            latents=latents,
+            negative_prompt=negative_prior_prompt,
+            guidance_scale=guidance_scale,
+        )
+        zero_image_emb = out_zero.negative_image_embeds if negative_prompt == "" else out_zero.image_embeds
+
+        return KandinskyPriorPipelineOutput(image_embeds=image_emb, negative_image_embeds=zero_image_emb)
+
+    # Copied from diffusers.pipelines.unclip.pipeline_unclip.UnCLIPPipeline.prepare_latents
+    def prepare_latents(self, shape, dtype, device, generator, latents, scheduler):
+        if latents is None:
+            latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        else:
+            if latents.shape != shape:
+                raise ValueError(f"Unexpected latents shape, got {latents.shape}, expected {shape}")
+            latents = latents.to(device)
+
+        latents = latents * scheduler.init_noise_sigma
+        return latents
+
+    def get_zero_embed(self, batch_size=1, device=None):
+        device = device or self.device
+        zero_img = torch.zeros(1, 3, self.image_encoder.config.image_size, self.image_encoder.config.image_size).to(
+            device=device, dtype=self.image_encoder.dtype
+        )
+        zero_image_emb = self.image_encoder(zero_img)["image_embeds"]
+        zero_image_emb = zero_image_emb.repeat(batch_size, 1)
+        return zero_image_emb
+
+    def _encode_prompt(
+        self,
+        prompt,
+        device,
+        num_images_per_prompt,
+        do_classifier_free_guidance,
+        negative_prompt=None,
+    ):
+        batch_size = len(prompt) if isinstance(prompt, list) else 1
+        # get prompt text embeddings
+        text_inputs = self.tokenizer(
+            prompt,
+            padding="max_length",
+            max_length=self.tokenizer.model_max_length,
+            truncation=True,
+            return_tensors="pt",
+        )
+        text_input_ids = text_inputs.input_ids
+        text_mask = text_inputs.attention_mask.bool().to(device)
+
+        untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+        if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(text_input_ids, untruncated_ids):
+            removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1])
+            logger.warning(
+                "The following part of your input was truncated because CLIP can only handle sequences up to"
+                f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+            )
+            text_input_ids = text_input_ids[:, : self.tokenizer.model_max_length]
+
+        text_encoder_output = self.text_encoder(text_input_ids.to(device))
+
+        prompt_embeds = text_encoder_output.text_embeds
+        text_encoder_hidden_states = text_encoder_output.last_hidden_state
+
+        prompt_embeds = prompt_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+        text_encoder_hidden_states = text_encoder_hidden_states.repeat_interleave(num_images_per_prompt, dim=0)
+        text_mask = text_mask.repeat_interleave(num_images_per_prompt, dim=0)
+
+        if do_classifier_free_guidance:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=self.tokenizer.model_max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+            uncond_text_mask = uncond_input.attention_mask.bool().to(device)
+            negative_prompt_embeds_text_encoder_output = self.text_encoder(uncond_input.input_ids.to(device))
+
+            negative_prompt_embeds = negative_prompt_embeds_text_encoder_output.text_embeds
+            uncond_text_encoder_hidden_states = negative_prompt_embeds_text_encoder_output.last_hidden_state
+
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+
+            seq_len = negative_prompt_embeds.shape[1]
+            negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt)
+            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len)
+
+            seq_len = uncond_text_encoder_hidden_states.shape[1]
+            uncond_text_encoder_hidden_states = uncond_text_encoder_hidden_states.repeat(1, num_images_per_prompt, 1)
+            uncond_text_encoder_hidden_states = uncond_text_encoder_hidden_states.view(
+                batch_size * num_images_per_prompt, seq_len, -1
+            )
+            uncond_text_mask = uncond_text_mask.repeat_interleave(num_images_per_prompt, dim=0)
+
+            # done duplicates
+
+            # For classifier free guidance, we need to do two forward passes.
+            # Here we concatenate the unconditional and text embeddings into a single batch
+            # to avoid doing two forward passes
+            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
+            text_encoder_hidden_states = torch.cat([uncond_text_encoder_hidden_states, text_encoder_hidden_states])
+
+            text_mask = torch.cat([uncond_text_mask, text_mask])
+
+        return prompt_embeds, text_encoder_hidden_states, text_mask
+
+    @torch.no_grad()
+    @replace_example_docstring(EXAMPLE_DOC_STRING)
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        num_images_per_prompt: int = 1,
+        num_inference_steps: int = 25,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        guidance_scale: float = 4.0,
+        output_type: Optional[str] = "pt",
+        return_dict: bool = True,
+    ):
+        """
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`):
+                The prompt or prompts to guide the image generation.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
+                if `guidance_scale` is less than `1`).
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            num_inference_steps (`int`, *optional*, defaults to 25):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            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.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 be generated by sampling using the supplied random `generator`.
+            guidance_scale (`float`, *optional*, defaults to 4.0):
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
+            output_type (`str`, *optional*, defaults to `"pt"`):
+                The output format of the generate image. Choose between: `"np"` (`np.array`) or `"pt"`
+                (`torch.Tensor`).
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
+
+        Examples:
+
+        Returns:
+            [`KandinskyPriorPipelineOutput`] or `tuple`
+        """
+
+        if isinstance(prompt, str):
+            prompt = [prompt]
+        elif not isinstance(prompt, list):
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        if isinstance(negative_prompt, str):
+            negative_prompt = [negative_prompt]
+        elif not isinstance(negative_prompt, list) and negative_prompt is not None:
+            raise ValueError(f"`negative_prompt` has to be of type `str` or `list` but is {type(negative_prompt)}")
+
+        # if the negative prompt is defined we double the batch size to
+        # directly retrieve the negative prompt embedding
+        if negative_prompt is not None:
+            prompt = prompt + negative_prompt
+            negative_prompt = 2 * negative_prompt
+
+        device = self._execution_device
+
+        batch_size = len(prompt)
+        batch_size = batch_size * num_images_per_prompt
+
+        do_classifier_free_guidance = guidance_scale > 1.0
+        prompt_embeds, text_encoder_hidden_states, text_mask = self._encode_prompt(
+            prompt, device, num_images_per_prompt, do_classifier_free_guidance, negative_prompt
+        )
+
+        # prior
+        self.scheduler.set_timesteps(num_inference_steps, device=device)
+        prior_timesteps_tensor = self.scheduler.timesteps
+
+        embedding_dim = self.prior.config.embedding_dim
+
+        latents = self.prepare_latents(
+            (batch_size, embedding_dim),
+            prompt_embeds.dtype,
+            device,
+            generator,
+            latents,
+            self.scheduler,
+        )
+
+        for i, t in enumerate(self.progress_bar(prior_timesteps_tensor)):
+            # expand the latents if we are doing classifier free guidance
+            latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
+
+            predicted_image_embedding = self.prior(
+                latent_model_input,
+                timestep=t,
+                proj_embedding=prompt_embeds,
+                encoder_hidden_states=text_encoder_hidden_states,
+                attention_mask=text_mask,
+            ).predicted_image_embedding
+
+            if do_classifier_free_guidance:
+                predicted_image_embedding_uncond, predicted_image_embedding_text = predicted_image_embedding.chunk(2)
+                predicted_image_embedding = predicted_image_embedding_uncond + guidance_scale * (
+                    predicted_image_embedding_text - predicted_image_embedding_uncond
+                )
+
+            if i + 1 == prior_timesteps_tensor.shape[0]:
+                prev_timestep = None
+            else:
+                prev_timestep = prior_timesteps_tensor[i + 1]
+
+            latents = self.scheduler.step(
+                predicted_image_embedding,
+                timestep=t,
+                sample=latents,
+                generator=generator,
+                prev_timestep=prev_timestep,
+            ).prev_sample
+
+            if XLA_AVAILABLE:
+                xm.mark_step()
+
+        latents = self.prior.post_process_latents(latents)
+
+        image_embeddings = latents
+
+        # if negative prompt has been defined, we retrieve split the image embedding into two
+        if negative_prompt is None:
+            zero_embeds = self.get_zero_embed(latents.shape[0], device=latents.device)
+
+            self.maybe_free_model_hooks()
+        else:
+            image_embeddings, zero_embeds = image_embeddings.chunk(2)
+
+            if hasattr(self, "final_offload_hook") and self.final_offload_hook is not None:
+                self.prior_hook.offload()
+
+        if output_type not in ["pt", "np"]:
+            raise ValueError(f"Only the output types `pt` and `np` are supported not output_type={output_type}")
+
+        if output_type == "np":
+            image_embeddings = image_embeddings.cpu().numpy()
+            zero_embeds = zero_embeds.cpu().numpy()
+
+        if not return_dict:
+            return (image_embeddings, zero_embeds)
+
+        return KandinskyPriorPipelineOutput(image_embeds=image_embeddings, negative_image_embeds=zero_embeds)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/kandinsky/text_encoder.py

@@ -0,0 +1,27 @@
+import torch
+from transformers import PreTrainedModel, XLMRobertaConfig, XLMRobertaModel
+
+
+class MCLIPConfig(XLMRobertaConfig):
+    model_type = "M-CLIP"
+
+    def __init__(self, transformerDimSize=1024, imageDimSize=768, **kwargs):
+        self.transformerDimensions = transformerDimSize
+        self.numDims = imageDimSize
+        super().__init__(**kwargs)
+
+
+class MultilingualCLIP(PreTrainedModel):
+    config_class = MCLIPConfig
+
+    def __init__(self, config, *args, **kwargs):
+        super().__init__(config, *args, **kwargs)
+        self.transformer = XLMRobertaModel(config)
+        self.LinearTransformation = torch.nn.Linear(
+            in_features=config.transformerDimensions, out_features=config.numDims
+        )
+
+    def forward(self, input_ids, attention_mask):
+        embs = self.transformer(input_ids=input_ids, attention_mask=attention_mask)[0]
+        embs2 = (embs * attention_mask.unsqueeze(2)).sum(dim=1) / attention_mask.sum(dim=1)[:, None]
+        return self.LinearTransformation(embs2), embs

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/semantic_stable_diffusion/__init__.py

@@ -0,0 +1,49 @@
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    DIFFUSERS_SLOW_IMPORT,
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    get_objects_from_module,
+    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_output"] = ["SemanticStableDiffusionPipelineOutput"]
+    _import_structure["pipeline_semantic_stable_diffusion"] = ["SemanticStableDiffusionPipeline"]
+
+
+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_semantic_stable_diffusion import SemanticStableDiffusionPipeline
+
+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)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/semantic_stable_diffusion/pipeline_output.py

@@ -0,0 +1,25 @@
+from dataclasses import dataclass
+from typing import List, Optional, Union
+
+import numpy as np
+import PIL.Image
+
+from ...utils import BaseOutput
+
+
+@dataclass
+class SemanticStableDiffusionPipelineOutput(BaseOutput):
+    """
+    Output class for Stable Diffusion pipelines.
+
+    Args:
+        images (`List[PIL.Image.Image]` or `np.ndarray`)
+            List of denoised PIL images of length `batch_size` or NumPy array of shape `(batch_size, height, width,
+            num_channels)`.
+        nsfw_content_detected (`List[bool]`)
+            List indicating whether the corresponding generated image contains “not-safe-for-work” (nsfw) content or
+            `None` if safety checking could not be performed.
+    """
+
+    images: Union[List[PIL.Image.Image], np.ndarray]
+    nsfw_content_detected: Optional[List[bool]]

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/semantic_stable_diffusion/pipeline_semantic_stable_diffusion.py

@@ -0,0 +1,733 @@
+import inspect
+from itertools import repeat
+from typing import Callable, List, Optional, Union
+
+import torch
+from transformers import CLIPImageProcessor, CLIPTextModel, CLIPTokenizer
+
+from ...image_processor import VaeImageProcessor
+from ...models import AutoencoderKL, UNet2DConditionModel
+from ...pipelines.stable_diffusion.safety_checker import StableDiffusionSafetyChecker
+from ...schedulers import KarrasDiffusionSchedulers
+from ...utils import deprecate, is_torch_xla_available, logging
+from ...utils.torch_utils import randn_tensor
+from ..pipeline_utils import DeprecatedPipelineMixin, DiffusionPipeline, StableDiffusionMixin
+from .pipeline_output import SemanticStableDiffusionPipelineOutput
+
+
+if is_torch_xla_available():
+    import torch_xla.core.xla_model as xm
+
+    XLA_AVAILABLE = True
+else:
+    XLA_AVAILABLE = False
+
+logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
+
+
+class SemanticStableDiffusionPipeline(DeprecatedPipelineMixin, DiffusionPipeline, StableDiffusionMixin):
+    _last_supported_version = "0.33.1"
+    r"""
+    Pipeline for text-to-image generation using Stable Diffusion with latent editing.
+
+    This model inherits from [`DiffusionPipeline`] and builds on the [`StableDiffusionPipeline`]. Check the superclass
+    documentation for the generic methods implemented for all pipelines (downloading, saving, running on a particular
+    device, etc.).
+
+    Args:
+        vae ([`AutoencoderKL`]):
+            Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations.
+        text_encoder ([`~transformers.CLIPTextModel`]):
+            Frozen text-encoder ([clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14)).
+        tokenizer ([`~transformers.CLIPTokenizer`]):
+            A `CLIPTokenizer` to tokenize text.
+        unet ([`UNet2DConditionModel`]):
+            A `UNet2DConditionModel` to denoise the encoded image latents.
+        scheduler ([`SchedulerMixin`]):
+            A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of
+            [`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`].
+        safety_checker ([`Q16SafetyChecker`]):
+            Classification module that estimates whether generated images could be considered offensive or harmful.
+            Please refer to the [model card](https://huggingface.co/runwayml/stable-diffusion-v1-5) for more details
+            about a model's potential harms.
+        feature_extractor ([`~transformers.CLIPImageProcessor`]):
+            A `CLIPImageProcessor` to extract features from generated images; used as inputs to the `safety_checker`.
+    """
+
+    model_cpu_offload_seq = "text_encoder->unet->vae"
+    _optional_components = ["safety_checker", "feature_extractor"]
+
+    def __init__(
+        self,
+        vae: AutoencoderKL,
+        text_encoder: CLIPTextModel,
+        tokenizer: CLIPTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: KarrasDiffusionSchedulers,
+        safety_checker: StableDiffusionSafetyChecker,
+        feature_extractor: CLIPImageProcessor,
+        requires_safety_checker: bool = True,
+    ):
+        super().__init__()
+
+        if safety_checker is None and requires_safety_checker:
+            logger.warning(
+                f"You have disabled the safety checker for {self.__class__} by passing `safety_checker=None`. Ensure"
+                " that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered"
+                " results in services or applications open to the public. Both the diffusers team and Hugging Face"
+                " strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling"
+                " it only for use-cases that involve analyzing network behavior or auditing its results. For more"
+                " information, please have a look at https://github.com/huggingface/diffusers/pull/254 ."
+            )
+
+        if safety_checker is not None and feature_extractor is None:
+            raise ValueError(
+                "Make sure to define a feature extractor when loading {self.__class__} if you want to use the safety"
+                " checker. If you do not want to use the safety checker, you can pass `'safety_checker=None'` instead."
+            )
+
+        self.register_modules(
+            vae=vae,
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            safety_checker=safety_checker,
+            feature_extractor=feature_extractor,
+        )
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1) if getattr(self, "vae", None) else 8
+        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self.register_to_config(requires_safety_checker=requires_safety_checker)
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.run_safety_checker
+    def run_safety_checker(self, image, device, dtype):
+        if self.safety_checker is None:
+            has_nsfw_concept = None
+        else:
+            if torch.is_tensor(image):
+                feature_extractor_input = self.image_processor.postprocess(image, output_type="pil")
+            else:
+                feature_extractor_input = self.image_processor.numpy_to_pil(image)
+            safety_checker_input = self.feature_extractor(feature_extractor_input, return_tensors="pt").to(device)
+            image, has_nsfw_concept = self.safety_checker(
+                images=image, clip_input=safety_checker_input.pixel_values.to(dtype)
+            )
+        return image, has_nsfw_concept
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.decode_latents
+    def decode_latents(self, latents):
+        deprecation_message = "The decode_latents method is deprecated and will be removed in 1.0.0. Please use VaeImageProcessor.postprocess(...) instead"
+        deprecate("decode_latents", "1.0.0", deprecation_message, standard_warn=False)
+
+        latents = 1 / self.vae.config.scaling_factor * latents
+        image = self.vae.decode(latents, return_dict=False)[0]
+        image = (image / 2 + 0.5).clamp(0, 1)
+        # we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
+        image = image.cpu().permute(0, 2, 3, 1).float().numpy()
+        return image
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_extra_step_kwargs
+    def prepare_extra_step_kwargs(self, generator, eta):
+        # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
+        # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
+        # eta corresponds to η in DDIM paper: https://huggingface.co/papers/2010.02502
+        # and should be between [0, 1]
+
+        accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        extra_step_kwargs = {}
+        if accepts_eta:
+            extra_step_kwargs["eta"] = eta
+
+        # check if the scheduler accepts generator
+        accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        if accepts_generator:
+            extra_step_kwargs["generator"] = generator
+        return extra_step_kwargs
+
+    # Copied from diffusers.pipelines.stable_diffusion_k_diffusion.pipeline_stable_diffusion_k_diffusion.StableDiffusionKDiffusionPipeline.check_inputs
+    def check_inputs(
+        self,
+        prompt,
+        height,
+        width,
+        callback_steps,
+        negative_prompt=None,
+        prompt_embeds=None,
+        negative_prompt_embeds=None,
+        callback_on_step_end_tensor_inputs=None,
+    ):
+        if height % 8 != 0 or width % 8 != 0:
+            raise ValueError(f"`height` and `width` have to be divisible by 8 but are {height} and {width}.")
+
+        if callback_steps is not None and (not isinstance(callback_steps, int) or callback_steps <= 0):
+            raise ValueError(
+                f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
+                f" {type(callback_steps)}."
+            )
+        if callback_on_step_end_tensor_inputs is not None and not all(
+            k in self._callback_tensor_inputs for k in callback_on_step_end_tensor_inputs
+        ):
+            raise ValueError(
+                f"`callback_on_step_end_tensor_inputs` has to be in {self._callback_tensor_inputs}, but found {[k for k in callback_on_step_end_tensor_inputs if k not in self._callback_tensor_inputs]}"
+            )
+
+        if prompt is not None and prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
+                " only forward one of the two."
+            )
+        elif prompt is None and prompt_embeds is None:
+            raise ValueError(
+                "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
+            )
+        elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        if negative_prompt is not None and negative_prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
+                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
+            )
+
+        if prompt_embeds is not None and negative_prompt_embeds is not None:
+            if prompt_embeds.shape != negative_prompt_embeds.shape:
+                raise ValueError(
+                    "`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
+                    f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
+                    f" {negative_prompt_embeds.shape}."
+                )
+
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_latents
+    def prepare_latents(self, batch_size, num_channels_latents, height, width, dtype, device, generator, latents=None):
+        shape = (
+            batch_size,
+            num_channels_latents,
+            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"
+                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+            )
+
+        if latents is None:
+            latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        else:
+            latents = latents.to(device)
+
+        # scale the initial noise by the standard deviation required by the scheduler
+        latents = latents * self.scheduler.init_noise_sigma
+        return latents
+
+    @torch.no_grad()
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        height: Optional[int] = None,
+        width: Optional[int] = None,
+        num_inference_steps: int = 50,
+        guidance_scale: float = 7.5,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
+        num_images_per_prompt: int = 1,
+        eta: float = 0.0,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: Optional[str] = "pil",
+        return_dict: bool = True,
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
+        callback_steps: int = 1,
+        editing_prompt: Optional[Union[str, List[str]]] = None,
+        editing_prompt_embeddings: Optional[torch.Tensor] = None,
+        reverse_editing_direction: Optional[Union[bool, List[bool]]] = False,
+        edit_guidance_scale: Optional[Union[float, List[float]]] = 5,
+        edit_warmup_steps: Optional[Union[int, List[int]]] = 10,
+        edit_cooldown_steps: Optional[Union[int, List[int]]] = None,
+        edit_threshold: Optional[Union[float, List[float]]] = 0.9,
+        edit_momentum_scale: Optional[float] = 0.1,
+        edit_mom_beta: Optional[float] = 0.4,
+        edit_weights: Optional[List[float]] = None,
+        sem_guidance: Optional[List[torch.Tensor]] = None,
+    ):
+        r"""
+        The call function to the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`):
+                The prompt or prompts to guide image generation.
+            height (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`):
+                The height in pixels of the generated image.
+            width (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`):
+                The width in pixels of the generated image.
+            num_inference_steps (`int`, *optional*, defaults to 50):
+                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+                expense of slower inference.
+            guidance_scale (`float`, *optional*, defaults to 7.5):
+                A higher guidance scale value encourages the model to generate images closely linked to the text
+                `prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide what to not include in image generation. If not defined, you need to
+                pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`).
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            eta (`float`, *optional*, defaults to 0.0):
+                Corresponds to parameter eta (η) from the [DDIM](https://huggingface.co/papers/2010.02502) paper. Only
+                applies to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers.
+            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.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 is generated by sampling using the supplied random `generator`.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generated image. Choose between `PIL.Image` or `np.array`.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] instead of a
+                plain tuple.
+            callback (`Callable`, *optional*):
+                A function that calls every `callback_steps` steps during inference. The function is 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 is called. If not specified, the callback is called at
+                every step.
+            editing_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to use for semantic guidance. Semantic guidance is disabled by setting
+                `editing_prompt = None`. Guidance direction of prompt should be specified via
+                `reverse_editing_direction`.
+            editing_prompt_embeddings (`torch.Tensor`, *optional*):
+                Pre-computed embeddings to use for semantic guidance. Guidance direction of embedding should be
+                specified via `reverse_editing_direction`.
+            reverse_editing_direction (`bool` or `List[bool]`, *optional*, defaults to `False`):
+                Whether the corresponding prompt in `editing_prompt` should be increased or decreased.
+            edit_guidance_scale (`float` or `List[float]`, *optional*, defaults to 5):
+                Guidance scale for semantic guidance. If provided as a list, values should correspond to
+                `editing_prompt`.
+            edit_warmup_steps (`float` or `List[float]`, *optional*, defaults to 10):
+                Number of diffusion steps (for each prompt) for which semantic guidance is not applied. Momentum is
+                calculated for those steps and applied once all warmup periods are over.
+            edit_cooldown_steps (`float` or `List[float]`, *optional*, defaults to `None`):
+                Number of diffusion steps (for each prompt) after which semantic guidance is longer applied.
+            edit_threshold (`float` or `List[float]`, *optional*, defaults to 0.9):
+                Threshold of semantic guidance.
+            edit_momentum_scale (`float`, *optional*, defaults to 0.1):
+                Scale of the momentum to be added to the semantic guidance at each diffusion step. If set to 0.0,
+                momentum is disabled. Momentum is already built up during warmup (for diffusion steps smaller than
+                `sld_warmup_steps`). Momentum is only added to latent guidance once all warmup periods are finished.
+            edit_mom_beta (`float`, *optional*, defaults to 0.4):
+                Defines how semantic guidance momentum builds up. `edit_mom_beta` indicates how much of the previous
+                momentum is kept. Momentum is already built up during warmup (for diffusion steps smaller than
+                `edit_warmup_steps`).
+            edit_weights (`List[float]`, *optional*, defaults to `None`):
+                Indicates how much each individual concept should influence the overall guidance. If no weights are
+                provided all concepts are applied equally.
+            sem_guidance (`List[torch.Tensor]`, *optional*):
+                List of pre-generated guidance vectors to be applied at generation. Length of the list has to
+                correspond to `num_inference_steps`.
+
+        Examples:
+
+        ```py
+        >>> import torch
+        >>> from diffusers import SemanticStableDiffusionPipeline
+
+        >>> pipe = SemanticStableDiffusionPipeline.from_pretrained(
+        ...     "runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16
+        ... )
+        >>> pipe = pipe.to("cuda")
+
+        >>> out = pipe(
+        ...     prompt="a photo of the face of a woman",
+        ...     num_images_per_prompt=1,
+        ...     guidance_scale=7,
+        ...     editing_prompt=[
+        ...         "smiling, smile",  # Concepts to apply
+        ...         "glasses, wearing glasses",
+        ...         "curls, wavy hair, curly hair",
+        ...         "beard, full beard, mustache",
+        ...     ],
+        ...     reverse_editing_direction=[
+        ...         False,
+        ...         False,
+        ...         False,
+        ...         False,
+        ...     ],  # Direction of guidance i.e. increase all concepts
+        ...     edit_warmup_steps=[10, 10, 10, 10],  # Warmup period for each concept
+        ...     edit_guidance_scale=[4, 5, 5, 5.4],  # Guidance scale for each concept
+        ...     edit_threshold=[
+        ...         0.99,
+        ...         0.975,
+        ...         0.925,
+        ...         0.96,
+        ...     ],  # Threshold for each concept. Threshold equals the percentile of the latent space that will be discarded. I.e. threshold=0.99 uses 1% of the latent dimensions
+        ...     edit_momentum_scale=0.3,  # Momentum scale that will be added to the latent guidance
+        ...     edit_mom_beta=0.6,  # Momentum beta
+        ...     edit_weights=[1, 1, 1, 1, 1],  # Weights of the individual concepts against each other
+        ... )
+        >>> image = out.images[0]
+        ```
+
+        Returns:
+            [`~pipelines.semantic_stable_diffusion.SemanticStableDiffusionPipelineOutput`] or `tuple`:
+                If `return_dict` is `True`,
+                [`~pipelines.semantic_stable_diffusion.SemanticStableDiffusionPipelineOutput`] is returned, otherwise a
+                `tuple` is returned where the first element is a list with the generated images and the second element
+                is a list of `bool`s indicating whether the corresponding generated image contains "not-safe-for-work"
+                (nsfw) content.
+        """
+        # 0. Default height and width to unet
+        height = height or self.unet.config.sample_size * self.vae_scale_factor
+        width = width or self.unet.config.sample_size * self.vae_scale_factor
+
+        # 1. Check inputs. Raise error if not correct
+        self.check_inputs(prompt, height, width, callback_steps)
+
+        # 2. Define call parameters
+        batch_size = 1 if isinstance(prompt, str) else len(prompt)
+        device = self._execution_device
+
+        if editing_prompt:
+            enable_edit_guidance = True
+            if isinstance(editing_prompt, str):
+                editing_prompt = [editing_prompt]
+            enabled_editing_prompts = len(editing_prompt)
+        elif editing_prompt_embeddings is not None:
+            enable_edit_guidance = True
+            enabled_editing_prompts = editing_prompt_embeddings.shape[0]
+        else:
+            enabled_editing_prompts = 0
+            enable_edit_guidance = False
+
+        # get prompt text embeddings
+        text_inputs = self.tokenizer(
+            prompt,
+            padding="max_length",
+            max_length=self.tokenizer.model_max_length,
+            return_tensors="pt",
+        )
+        text_input_ids = text_inputs.input_ids
+
+        if text_input_ids.shape[-1] > self.tokenizer.model_max_length:
+            removed_text = self.tokenizer.batch_decode(text_input_ids[:, self.tokenizer.model_max_length :])
+            logger.warning(
+                "The following part of your input was truncated because CLIP can only handle sequences up to"
+                f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+            )
+            text_input_ids = text_input_ids[:, : self.tokenizer.model_max_length]
+        text_embeddings = self.text_encoder(text_input_ids.to(device))[0]
+
+        # duplicate text embeddings for each generation per prompt, using mps friendly method
+        bs_embed, seq_len, _ = text_embeddings.shape
+        text_embeddings = text_embeddings.repeat(1, num_images_per_prompt, 1)
+        text_embeddings = text_embeddings.view(bs_embed * num_images_per_prompt, seq_len, -1)
+
+        if enable_edit_guidance:
+            # get safety text embeddings
+            if editing_prompt_embeddings is None:
+                edit_concepts_input = self.tokenizer(
+                    [x for item in editing_prompt for x in repeat(item, batch_size)],
+                    padding="max_length",
+                    max_length=self.tokenizer.model_max_length,
+                    return_tensors="pt",
+                )
+
+                edit_concepts_input_ids = edit_concepts_input.input_ids
+
+                if edit_concepts_input_ids.shape[-1] > self.tokenizer.model_max_length:
+                    removed_text = self.tokenizer.batch_decode(
+                        edit_concepts_input_ids[:, self.tokenizer.model_max_length :]
+                    )
+                    logger.warning(
+                        "The following part of your input was truncated because CLIP can only handle sequences up to"
+                        f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+                    )
+                    edit_concepts_input_ids = edit_concepts_input_ids[:, : self.tokenizer.model_max_length]
+                edit_concepts = self.text_encoder(edit_concepts_input_ids.to(device))[0]
+            else:
+                edit_concepts = editing_prompt_embeddings.to(device).repeat(batch_size, 1, 1)
+
+            # duplicate text embeddings for each generation per prompt, using mps friendly method
+            bs_embed_edit, seq_len_edit, _ = edit_concepts.shape
+            edit_concepts = edit_concepts.repeat(1, num_images_per_prompt, 1)
+            edit_concepts = edit_concepts.view(bs_embed_edit * num_images_per_prompt, seq_len_edit, -1)
+
+        # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
+        # of the Imagen paper: https://huggingface.co/papers/2205.11487 . `guidance_scale = 1`
+        # corresponds to doing no classifier free guidance.
+        do_classifier_free_guidance = guidance_scale > 1.0
+        # get unconditional embeddings for classifier free guidance
+
+        if do_classifier_free_guidance:
+            uncond_tokens: List[str]
+            if negative_prompt is None:
+                uncond_tokens = [""] * batch_size
+            elif type(prompt) is not type(negative_prompt):
+                raise TypeError(
+                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+                    f" {type(prompt)}."
+                )
+            elif isinstance(negative_prompt, str):
+                uncond_tokens = [negative_prompt]
+            elif batch_size != len(negative_prompt):
+                raise ValueError(
+                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+                    " the batch size of `prompt`."
+                )
+            else:
+                uncond_tokens = negative_prompt
+
+            max_length = text_input_ids.shape[-1]
+            uncond_input = self.tokenizer(
+                uncond_tokens,
+                padding="max_length",
+                max_length=max_length,
+                truncation=True,
+                return_tensors="pt",
+            )
+            uncond_embeddings = self.text_encoder(uncond_input.input_ids.to(device))[0]
+
+            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
+            seq_len = uncond_embeddings.shape[1]
+            uncond_embeddings = uncond_embeddings.repeat(1, num_images_per_prompt, 1)
+            uncond_embeddings = uncond_embeddings.view(batch_size * num_images_per_prompt, seq_len, -1)
+
+            # For classifier free guidance, we need to do two forward passes.
+            # Here we concatenate the unconditional and text embeddings into a single batch
+            # to avoid doing two forward passes
+            if enable_edit_guidance:
+                text_embeddings = torch.cat([uncond_embeddings, text_embeddings, edit_concepts])
+            else:
+                text_embeddings = torch.cat([uncond_embeddings, text_embeddings])
+        # get the initial random noise unless the user supplied it
+
+        # 4. Prepare timesteps
+        self.scheduler.set_timesteps(num_inference_steps, device=device)
+        timesteps = self.scheduler.timesteps
+
+        # 5. Prepare latent variables
+        num_channels_latents = self.unet.config.in_channels
+        latents = self.prepare_latents(
+            batch_size * num_images_per_prompt,
+            num_channels_latents,
+            height,
+            width,
+            text_embeddings.dtype,
+            device,
+            generator,
+            latents,
+        )
+
+        # 6. Prepare extra step kwargs.
+        extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)
+
+        # Initialize edit_momentum to None
+        edit_momentum = None
+
+        self.uncond_estimates = None
+        self.text_estimates = None
+        self.edit_estimates = None
+        self.sem_guidance = None
+
+        for i, t in enumerate(self.progress_bar(timesteps)):
+            # expand the latents if we are doing classifier free guidance
+            latent_model_input = (
+                torch.cat([latents] * (2 + enabled_editing_prompts)) if do_classifier_free_guidance else latents
+            )
+            latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
+
+            # predict the noise residual
+            noise_pred = self.unet(latent_model_input, t, encoder_hidden_states=text_embeddings).sample
+
+            # perform guidance
+            if do_classifier_free_guidance:
+                noise_pred_out = noise_pred.chunk(2 + enabled_editing_prompts)  # [b,4, 64, 64]
+                noise_pred_uncond, noise_pred_text = noise_pred_out[0], noise_pred_out[1]
+                noise_pred_edit_concepts = noise_pred_out[2:]
+
+                # default text guidance
+                noise_guidance = guidance_scale * (noise_pred_text - noise_pred_uncond)
+                # noise_guidance = (noise_pred_text - noise_pred_edit_concepts[0])
+
+                if self.uncond_estimates is None:
+                    self.uncond_estimates = torch.zeros((num_inference_steps + 1, *noise_pred_uncond.shape))
+                self.uncond_estimates[i] = noise_pred_uncond.detach().cpu()
+
+                if self.text_estimates is None:
+                    self.text_estimates = torch.zeros((num_inference_steps + 1, *noise_pred_text.shape))
+                self.text_estimates[i] = noise_pred_text.detach().cpu()
+
+                if self.edit_estimates is None and enable_edit_guidance:
+                    self.edit_estimates = torch.zeros(
+                        (num_inference_steps + 1, len(noise_pred_edit_concepts), *noise_pred_edit_concepts[0].shape)
+                    )
+
+                if self.sem_guidance is None:
+                    self.sem_guidance = torch.zeros((num_inference_steps + 1, *noise_pred_text.shape))
+
+                if edit_momentum is None:
+                    edit_momentum = torch.zeros_like(noise_guidance)
+
+                if enable_edit_guidance:
+                    concept_weights = torch.zeros(
+                        (len(noise_pred_edit_concepts), noise_guidance.shape[0]),
+                        device=device,
+                        dtype=noise_guidance.dtype,
+                    )
+                    noise_guidance_edit = torch.zeros(
+                        (len(noise_pred_edit_concepts), *noise_guidance.shape),
+                        device=device,
+                        dtype=noise_guidance.dtype,
+                    )
+                    # noise_guidance_edit = torch.zeros_like(noise_guidance)
+                    warmup_inds = []
+                    for c, noise_pred_edit_concept in enumerate(noise_pred_edit_concepts):
+                        self.edit_estimates[i, c] = noise_pred_edit_concept
+                        if isinstance(edit_guidance_scale, list):
+                            edit_guidance_scale_c = edit_guidance_scale[c]
+                        else:
+                            edit_guidance_scale_c = edit_guidance_scale
+
+                        if isinstance(edit_threshold, list):
+                            edit_threshold_c = edit_threshold[c]
+                        else:
+                            edit_threshold_c = edit_threshold
+                        if isinstance(reverse_editing_direction, list):
+                            reverse_editing_direction_c = reverse_editing_direction[c]
+                        else:
+                            reverse_editing_direction_c = reverse_editing_direction
+                        if edit_weights:
+                            edit_weight_c = edit_weights[c]
+                        else:
+                            edit_weight_c = 1.0
+                        if isinstance(edit_warmup_steps, list):
+                            edit_warmup_steps_c = edit_warmup_steps[c]
+                        else:
+                            edit_warmup_steps_c = edit_warmup_steps
+
+                        if isinstance(edit_cooldown_steps, list):
+                            edit_cooldown_steps_c = edit_cooldown_steps[c]
+                        elif edit_cooldown_steps is None:
+                            edit_cooldown_steps_c = i + 1
+                        else:
+                            edit_cooldown_steps_c = edit_cooldown_steps
+                        if i >= edit_warmup_steps_c:
+                            warmup_inds.append(c)
+                        if i >= edit_cooldown_steps_c:
+                            noise_guidance_edit[c, :, :, :, :] = torch.zeros_like(noise_pred_edit_concept)
+                            continue
+
+                        noise_guidance_edit_tmp = noise_pred_edit_concept - noise_pred_uncond
+                        # tmp_weights = (noise_pred_text - noise_pred_edit_concept).sum(dim=(1, 2, 3))
+                        tmp_weights = (noise_guidance - noise_pred_edit_concept).sum(dim=(1, 2, 3))
+
+                        tmp_weights = torch.full_like(tmp_weights, edit_weight_c)  # * (1 / enabled_editing_prompts)
+                        if reverse_editing_direction_c:
+                            noise_guidance_edit_tmp = noise_guidance_edit_tmp * -1
+                        concept_weights[c, :] = tmp_weights
+
+                        noise_guidance_edit_tmp = noise_guidance_edit_tmp * edit_guidance_scale_c
+
+                        # torch.quantile function expects float32
+                        if noise_guidance_edit_tmp.dtype == torch.float32:
+                            tmp = torch.quantile(
+                                torch.abs(noise_guidance_edit_tmp).flatten(start_dim=2),
+                                edit_threshold_c,
+                                dim=2,
+                                keepdim=False,
+                            )
+                        else:
+                            tmp = torch.quantile(
+                                torch.abs(noise_guidance_edit_tmp).flatten(start_dim=2).to(torch.float32),
+                                edit_threshold_c,
+                                dim=2,
+                                keepdim=False,
+                            ).to(noise_guidance_edit_tmp.dtype)
+
+                        noise_guidance_edit_tmp = torch.where(
+                            torch.abs(noise_guidance_edit_tmp) >= tmp[:, :, None, None],
+                            noise_guidance_edit_tmp,
+                            torch.zeros_like(noise_guidance_edit_tmp),
+                        )
+                        noise_guidance_edit[c, :, :, :, :] = noise_guidance_edit_tmp
+
+                        # noise_guidance_edit = noise_guidance_edit + noise_guidance_edit_tmp
+
+                    warmup_inds = torch.tensor(warmup_inds).to(device)
+                    if len(noise_pred_edit_concepts) > warmup_inds.shape[0] > 0:
+                        concept_weights = concept_weights.to("cpu")  # Offload to cpu
+                        noise_guidance_edit = noise_guidance_edit.to("cpu")
+
+                        concept_weights_tmp = torch.index_select(concept_weights.to(device), 0, warmup_inds)
+                        concept_weights_tmp = torch.where(
+                            concept_weights_tmp < 0, torch.zeros_like(concept_weights_tmp), concept_weights_tmp
+                        )
+                        concept_weights_tmp = concept_weights_tmp / concept_weights_tmp.sum(dim=0)
+                        # concept_weights_tmp = torch.nan_to_num(concept_weights_tmp)
+
+                        noise_guidance_edit_tmp = torch.index_select(noise_guidance_edit.to(device), 0, warmup_inds)
+                        noise_guidance_edit_tmp = torch.einsum(
+                            "cb,cbijk->bijk", concept_weights_tmp, noise_guidance_edit_tmp
+                        )
+                        noise_guidance = noise_guidance + noise_guidance_edit_tmp
+
+                        self.sem_guidance[i] = noise_guidance_edit_tmp.detach().cpu()
+
+                        del noise_guidance_edit_tmp
+                        del concept_weights_tmp
+                        concept_weights = concept_weights.to(device)
+                        noise_guidance_edit = noise_guidance_edit.to(device)
+
+                    concept_weights = torch.where(
+                        concept_weights < 0, torch.zeros_like(concept_weights), concept_weights
+                    )
+
+                    concept_weights = torch.nan_to_num(concept_weights)
+
+                    noise_guidance_edit = torch.einsum("cb,cbijk->bijk", concept_weights, noise_guidance_edit)
+                    noise_guidance_edit = noise_guidance_edit.to(edit_momentum.device)
+
+                    noise_guidance_edit = noise_guidance_edit + edit_momentum_scale * edit_momentum
+
+                    edit_momentum = edit_mom_beta * edit_momentum + (1 - edit_mom_beta) * noise_guidance_edit
+
+                    if warmup_inds.shape[0] == len(noise_pred_edit_concepts):
+                        noise_guidance = noise_guidance + noise_guidance_edit
+                        self.sem_guidance[i] = noise_guidance_edit.detach().cpu()
+
+                if sem_guidance is not None:
+                    edit_guidance = sem_guidance[i].to(device)
+                    noise_guidance = noise_guidance + edit_guidance
+
+                noise_pred = noise_pred_uncond + noise_guidance
+
+                # compute the previous noisy sample x_t -> x_t-1
+            latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs).prev_sample
+
+            # call the callback, if provided
+            if callback is not None and i % callback_steps == 0:
+                step_idx = i // getattr(self.scheduler, "order", 1)
+                callback(step_idx, t, latents)
+
+            if XLA_AVAILABLE:
+                xm.mark_step()
+
+        # 8. Post-processing
+        if not output_type == "latent":
+            image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0]
+            image, has_nsfw_concept = self.run_safety_checker(image, device, text_embeddings.dtype)
+        else:
+            image = latents
+            has_nsfw_concept = None
+
+        if has_nsfw_concept is None:
+            do_denormalize = [True] * image.shape[0]
+        else:
+            do_denormalize = [not has_nsfw for has_nsfw in has_nsfw_concept]
+
+        image = self.image_processor.postprocess(image, output_type=output_type, do_denormalize=do_denormalize)
+
+        if not return_dict:
+            return (image, has_nsfw_concept)
+
+        return SemanticStableDiffusionPipelineOutput(images=image, nsfw_content_detected=has_nsfw_concept)

pythonProject/.venv/Lib/site-packages/diffusers/pipelines/shap_e/camera.py

@@ -0,0 +1,147 @@
+# Copyright 2025 Open AI and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from dataclasses import dataclass
+from typing import Tuple
+
+import numpy as np
+import torch
+
+
+@dataclass
+class DifferentiableProjectiveCamera:
+    """
+    Implements a batch, differentiable, standard pinhole camera
+    """
+
+    origin: torch.Tensor  # [batch_size x 3]
+    x: torch.Tensor  # [batch_size x 3]
+    y: torch.Tensor  # [batch_size x 3]
+    z: torch.Tensor  # [batch_size x 3]
+    width: int
+    height: int
+    x_fov: float
+    y_fov: float
+    shape: Tuple[int]
+
+    def __post_init__(self):
+        assert self.x.shape[0] == self.y.shape[0] == self.z.shape[0] == self.origin.shape[0]
+        assert self.x.shape[1] == self.y.shape[1] == self.z.shape[1] == self.origin.shape[1] == 3
+        assert len(self.x.shape) == len(self.y.shape) == len(self.z.shape) == len(self.origin.shape) == 2
+
+    def resolution(self):
+        return torch.from_numpy(np.array([self.width, self.height], dtype=np.float32))
+
+    def fov(self):
+        return torch.from_numpy(np.array([self.x_fov, self.y_fov], dtype=np.float32))
+
+    def get_image_coords(self) -> torch.Tensor:
+        """
+        :return: coords of shape (width * height, 2)
+        """
+        pixel_indices = torch.arange(self.height * self.width)
+        coords = torch.stack(
+            [
+                pixel_indices % self.width,
+                torch.div(pixel_indices, self.width, rounding_mode="trunc"),
+            ],
+            axis=1,
+        )
+        return coords
+
+    @property
+    def camera_rays(self):
+        batch_size, *inner_shape = self.shape
+        inner_batch_size = int(np.prod(inner_shape))
+
+        coords = self.get_image_coords()
+        coords = torch.broadcast_to(coords.unsqueeze(0), [batch_size * inner_batch_size, *coords.shape])
+        rays = self.get_camera_rays(coords)
+
+        rays = rays.view(batch_size, inner_batch_size * self.height * self.width, 2, 3)
+
+        return rays
+
+    def get_camera_rays(self, coords: torch.Tensor) -> torch.Tensor:
+        batch_size, *shape, n_coords = coords.shape
+        assert n_coords == 2
+        assert batch_size == self.origin.shape[0]
+
+        flat = coords.view(batch_size, -1, 2)
+
+        res = self.resolution()
+        fov = self.fov()
+
+        fracs = (flat.float() / (res - 1)) * 2 - 1
+        fracs = fracs * torch.tan(fov / 2)
+
+        fracs = fracs.view(batch_size, -1, 2)
+        directions = (
+            self.z.view(batch_size, 1, 3)
+            + self.x.view(batch_size, 1, 3) * fracs[:, :, :1]
+            + self.y.view(batch_size, 1, 3) * fracs[:, :, 1:]
+        )
+        directions = directions / directions.norm(dim=-1, keepdim=True)
+        rays = torch.stack(
+            [
+                torch.broadcast_to(self.origin.view(batch_size, 1, 3), [batch_size, directions.shape[1], 3]),
+                directions,
+            ],
+            dim=2,
+        )
+        return rays.view(batch_size, *shape, 2, 3)
+
+    def resize_image(self, width: int, height: int) -> "DifferentiableProjectiveCamera":
+        """
+        Creates a new camera for the resized view assuming the aspect ratio does not change.
+        """
+        assert width * self.height == height * self.width, "The aspect ratio should not change."
+        return DifferentiableProjectiveCamera(
+            origin=self.origin,
+            x=self.x,
+            y=self.y,
+            z=self.z,
+            width=width,
+            height=height,
+            x_fov=self.x_fov,
+            y_fov=self.y_fov,
+        )
+
+
+def create_pan_cameras(size: int) -> DifferentiableProjectiveCamera:
+    origins = []
+    xs = []
+    ys = []
+    zs = []
+    for theta in np.linspace(0, 2 * np.pi, num=20):
+        z = np.array([np.sin(theta), np.cos(theta), -0.5])
+        z /= np.sqrt(np.sum(z**2))
+        origin = -z * 4
+        x = np.array([np.cos(theta), -np.sin(theta), 0.0])
+        y = np.cross(z, x)
+        origins.append(origin)
+        xs.append(x)
+        ys.append(y)
+        zs.append(z)
+    return DifferentiableProjectiveCamera(
+        origin=torch.from_numpy(np.stack(origins, axis=0)).float(),
+        x=torch.from_numpy(np.stack(xs, axis=0)).float(),
+        y=torch.from_numpy(np.stack(ys, axis=0)).float(),
+        z=torch.from_numpy(np.stack(zs, axis=0)).float(),
+        width=size,
+        height=size,
+        x_fov=0.7,
+        y_fov=0.7,
+        shape=(1, len(xs)),
+    )