diffusers 0.33.0__py3-none-any.whl → 0.34.0__py3-none-any.whl

diffusers/utils/dummy_pt_objects.py

@@ -160,6 +160,21 @@ class AutoencoderKLCogVideoX(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class AutoencoderKLCosmos(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class AutoencoderKLHunyuanVideo(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -310,6 +325,21 @@ class CacheMixin(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class ChromaTransformer2DModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class CogVideoXTransformer3DModel(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -430,6 +460,21 @@ class ControlNetXSAdapter(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class CosmosTransformer3DModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class DiTTransformer2DModel(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -505,6 +550,21 @@ class FluxTransformer2DModel(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class HiDreamImageTransformer2DModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class HunyuanDiT2DControlNetModel(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -550,6 +610,21 @@ class HunyuanDiT2DMultiControlNetModel(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class HunyuanVideoFramepackTransformer3DModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class HunyuanVideoTransformer3DModel(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -775,6 +850,21 @@ class PriorTransformer(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class SanaControlNetModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class SanaTransformer2DModel(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -1075,6 +1165,21 @@ class WanTransformer3DModel(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class WanVACETransformer3DModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 def get_constant_schedule(*args, **kwargs):
     requires_backends(get_constant_schedule, ["torch"])
 
@@ -1733,6 +1838,21 @@ class FlowMatchHeunDiscreteScheduler(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class FlowMatchLCMScheduler(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class HeunDiscreteScheduler(metaclass=DummyObject):
     _backends = ["torch"]
 

diffusers/utils/dummy_torch_and_transformers_objects.py

@@ -272,6 +272,36 @@ class AuraFlowPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class ChromaImg2ImgPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
+class ChromaPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class CLIPImageProjection(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 
@@ -392,6 +422,81 @@ class CogView4Pipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class ConsisIDPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
+class Cosmos2TextToImagePipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
+class Cosmos2VideoToWorldPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
+class CosmosTextToWorldPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
+class CosmosVideoToWorldPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class CycleDiffusionPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 
@@ -617,6 +722,21 @@ class FluxPriorReduxPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class HiDreamImagePipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class HunyuanDiTControlNetPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 
@@ -677,6 +797,21 @@ class HunyuanSkyreelsImageToVideoPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class HunyuanVideoFramepackPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class HunyuanVideoImageToVideoPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 
@@ -1232,6 +1367,21 @@ class LTXImageToVideoPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class LTXLatentUpsamplePipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class LTXPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 
@@ -1487,6 +1637,21 @@ class ReduxImageEncoder(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class SanaControlNetPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class SanaPAGPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 
@@ -1517,6 +1682,21 @@ class SanaPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class SanaSprintImg2ImgPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class SanaSprintPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 
@@ -2702,6 +2882,36 @@ class VideoToVideoSDPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class VisualClozeGenerationPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
+class VisualClozePipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class VQDiffusionPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 
@@ -2747,6 +2957,21 @@ class WanPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class WanVACEPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class WanVideoToVideoPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 

diffusers/utils/dynamic_modules_utils.py

@@ -154,12 +154,30 @@ def check_imports(filename):
     return get_relative_imports(filename)
 
 
-def get_class_in_module(class_name, module_path):
+def get_class_in_module(class_name, module_path, pretrained_model_name_or_path=None):
     """
     Import a module on the cache directory for modules and extract a class from it.
     """
     module_path = module_path.replace(os.path.sep, ".")
-    module = importlib.import_module(module_path)
+    try:
+        module = importlib.import_module(module_path)
+    except ModuleNotFoundError as e:
+        # This can happen when the repo id contains ".", which Python's import machinery interprets as a directory
+        # separator. We do a bit of monkey patching to detect and fix this case.
+        if not (
+            pretrained_model_name_or_path is not None
+            and "." in pretrained_model_name_or_path
+            and module_path.startswith("diffusers_modules")
+            and pretrained_model_name_or_path.replace("/", "--") in module_path
+        ):
+            raise e  # We can't figure this one out, just reraise the original error
+
+        corrected_path = os.path.join(HF_MODULES_CACHE, module_path.replace(".", "/")) + ".py"
+        corrected_path = corrected_path.replace(
+            pretrained_model_name_or_path.replace("/", "--").replace(".", "/"),
+            pretrained_model_name_or_path.replace("/", "--"),
+        )
+        module = importlib.machinery.SourceFileLoader(module_path, corrected_path).load_module()
 
     if class_name is None:
         return find_pipeline_class(module)
@@ -454,4 +472,4 @@ def get_class_from_dynamic_module(
         revision=revision,
         local_files_only=local_files_only,
     )
-    return get_class_in_module(class_name, final_module.replace(".py", ""))
+    return get_class_in_module(class_name, final_module.replace(".py", ""), pretrained_model_name_or_path)

diffusers/utils/export_utils.py

@@ -155,7 +155,7 @@ def export_to_video(
     bitrate:
         Set a constant bitrate for the video encoding. Default is None causing `quality` parameter to be used instead.
         Better quality videos with smaller file sizes will result from using the `quality` variable bitrate parameter
-        rather than specifiying a fixed bitrate with this parameter.
+        rather than specifying a fixed bitrate with this parameter.
 
     macro_block_size:
         Size constraint for video. Width and height, must be divisible by this number. If not divisible by this number

diffusers/utils/import_utils.py

@@ -1,4 +1,4 @@
-# Copyright 2024 The HuggingFace Team. All rights reserved.
+# 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.
@@ -16,13 +16,14 @@ Import utilities: Utilities related to imports and our lazy inits.
 """
 
 import importlib.util
+import inspect
 import operator as op
 import os
 import sys
-from collections import OrderedDict
+from collections import OrderedDict, defaultdict
 from itertools import chain
 from types import ModuleType
-from typing import Any, Union
+from typing import Any, Tuple, Union
 
 from huggingface_hub.utils import is_jinja_available  # noqa: F401
 from packaging.version import Version, parse
@@ -35,7 +36,10 @@ if sys.version_info < (3, 8):
     import importlib_metadata
 else:
     import importlib.metadata as importlib_metadata
-
+try:
+    _package_map = importlib_metadata.packages_distributions()  # load-once to avoid expensive calls
+except Exception:
+    _package_map = None
 
 logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
 
@@ -54,12 +58,33 @@ STR_OPERATION_TO_FUNC = {">": op.gt, ">=": op.ge, "==": op.eq, "!=": op.ne, "<="
 _is_google_colab = "google.colab" in sys.modules or any(k.startswith("COLAB_") for k in os.environ)
 
 
-def _is_package_available(pkg_name: str):
+def _is_package_available(pkg_name: str, get_dist_name: bool = False) -> Tuple[bool, str]:
+    global _package_map
     pkg_exists = importlib.util.find_spec(pkg_name) is not None
     pkg_version = "N/A"
 
     if pkg_exists:
+        if _package_map is None:
+            _package_map = defaultdict(list)
+            try:
+                # Fallback for Python < 3.10
+                for dist in importlib_metadata.distributions():
+                    _top_level_declared = (dist.read_text("top_level.txt") or "").split()
+                    _infered_opt_names = {
+                        f.parts[0] if len(f.parts) > 1 else inspect.getmodulename(f) for f in (dist.files or [])
+                    } - {None}
+                    _top_level_inferred = filter(lambda name: "." not in name, _infered_opt_names)
+                    for pkg in _top_level_declared or _top_level_inferred:
+                        _package_map[pkg].append(dist.metadata["Name"])
+            except Exception as _:
+                pass
         try:
+            if get_dist_name and pkg_name in _package_map and _package_map[pkg_name]:
+                if len(_package_map[pkg_name]) > 1:
+                    logger.warning(
+                        f"Multiple distributions found for package {pkg_name}. Picked distribution: {_package_map[pkg_name][0]}"
+                    )
+                pkg_name = _package_map[pkg_name][0]
             pkg_version = importlib_metadata.version(pkg_name)
             logger.debug(f"Successfully imported {pkg_name} version {pkg_version}")
         except (ImportError, importlib_metadata.PackageNotFoundError):
@@ -74,6 +99,7 @@ if USE_TORCH in ENV_VARS_TRUE_AND_AUTO_VALUES and USE_TF not in ENV_VARS_TRUE_VA
 else:
     logger.info("Disabling PyTorch because USE_TORCH is set")
     _torch_available = False
+    _torch_version = "N/A"
 
 _jax_version = "N/A"
 _flax_version = "N/A"
@@ -101,18 +127,20 @@ _onnx_available = importlib.util.find_spec("onnxruntime") is not None
 if _onnx_available:
     candidates = (
         "onnxruntime",
+        "onnxruntime-cann",
+        "onnxruntime-directml",
+        "ort_nightly_directml",
         "onnxruntime-gpu",
         "ort_nightly_gpu",
-        "onnxruntime-directml",
+        "onnxruntime-migraphx",
         "onnxruntime-openvino",
-        "ort_nightly_directml",
+        "onnxruntime-qnn",
         "onnxruntime-rocm",
-        "onnxruntime-migraphx",
         "onnxruntime-training",
         "onnxruntime-vitisai",
     )
     _onnxruntime_version = None
-    # For the metadata, we have to look for both onnxruntime and onnxruntime-gpu
+    # For the metadata, we have to look for both onnxruntime and onnxruntime-x
     for pkg in candidates:
         try:
             _onnxruntime_version = importlib_metadata.version(pkg)
@@ -187,15 +215,11 @@ _xformers_available, _xformers_version = _is_package_available("xformers")
 _gguf_available, _gguf_version = _is_package_available("gguf")
 _torchao_available, _torchao_version = _is_package_available("torchao")
 _bitsandbytes_available, _bitsandbytes_version = _is_package_available("bitsandbytes")
-_torchao_available, _torchao_version = _is_package_available("torchao")
-
-_optimum_quanto_available = importlib.util.find_spec("optimum") is not None
-if _optimum_quanto_available:
-    try:
-        _optimum_quanto_version = importlib_metadata.version("optimum_quanto")
-        logger.debug(f"Successfully import optimum-quanto version {_optimum_quanto_version}")
-    except importlib_metadata.PackageNotFoundError:
-        _optimum_quanto_available = False
+_optimum_quanto_available, _optimum_quanto_version = _is_package_available("optimum", get_dist_name=True)
+_pytorch_retinaface_available, _pytorch_retinaface_version = _is_package_available("pytorch_retinaface")
+_better_profanity_available, _better_profanity_version = _is_package_available("better_profanity")
+_nltk_available, _nltk_version = _is_package_available("nltk")
+_cosmos_guardrail_available, _cosmos_guardrail_version = _is_package_available("cosmos_guardrail")
 
 
 def is_torch_available():
@@ -334,6 +358,26 @@ def is_timm_available():
     return _timm_available
 
 
+def is_pytorch_retinaface_available():
+    return _pytorch_retinaface_available
+
+
+def is_better_profanity_available():
+    return _better_profanity_available
+
+
+def is_nltk_available():
+    return _nltk_available
+
+
+def is_cosmos_guardrail_available():
+    return _cosmos_guardrail_available
+
+
+def is_hpu_available():
+    return all(importlib.util.find_spec(lib) for lib in ("habana_frameworks", "habana_frameworks.torch"))
+
+
 # docstyle-ignore
 FLAX_IMPORT_ERROR = """
 {0} requires the FLAX library but it was not found in your environment. Checkout the instructions on the
@@ -482,6 +526,22 @@ QUANTO_IMPORT_ERROR = """
 install optimum-quanto`
 """
 
+# docstyle-ignore
+PYTORCH_RETINAFACE_IMPORT_ERROR = """
+{0} requires the pytorch_retinaface library but it was not found in your environment. You can install it with pip: `pip install pytorch_retinaface`
+"""
+
+# docstyle-ignore
+BETTER_PROFANITY_IMPORT_ERROR = """
+{0} requires the better_profanity library but it was not found in your environment. You can install it with pip: `pip install better_profanity`
+"""
+
+# docstyle-ignore
+NLTK_IMPORT_ERROR = """
+{0} requires the nltk library but it was not found in your environment. You can install it with pip: `pip install nltk`
+"""
+
+
 BACKENDS_MAPPING = OrderedDict(
     [
         ("bs4", (is_bs4_available, BS4_IMPORT_ERROR)),
@@ -510,6 +570,9 @@ BACKENDS_MAPPING = OrderedDict(
         ("gguf", (is_gguf_available, GGUF_IMPORT_ERROR)),
         ("torchao", (is_torchao_available, TORCHAO_IMPORT_ERROR)),
         ("quanto", (is_optimum_quanto_available, QUANTO_IMPORT_ERROR)),
+        ("pytorch_retinaface", (is_pytorch_retinaface_available, PYTORCH_RETINAFACE_IMPORT_ERROR)),
+        ("better_profanity", (is_better_profanity_available, BETTER_PROFANITY_IMPORT_ERROR)),
+        ("nltk", (is_nltk_available, NLTK_IMPORT_ERROR)),
     ]
 )
 

diffusers/utils/logging.py

@@ -1,5 +1,5 @@
 # coding=utf-8
-# Copyright 2024 Optuna, Hugging Face
+# Copyright 2025 Optuna, Hugging Face
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.