Merge branch 'main' into remove-non-controlnet-modules

docs/source/en/api/models/controlnet_flux.md

@@ -42,4 +42,4 @@ pipe = FluxControlNetPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", co
 
 ## FluxControlNetOutput
 
-[[autodoc]] models.controlnet_flux.FluxControlNetOutput
+[[autodoc]] models.controlnets.controlnet_flux.FluxControlNetOutput

docs/source/en/api/models/controlnet_sparsectrl.md

@@ -43,4 +43,4 @@ controlnet = SparseControlNetModel.from_pretrained("guoyww/animatediff-sparsectr
 
 ## SparseControlNetOutput
 
-[[autodoc]] models.controlnet_sparsectrl.SparseControlNetOutput
+[[autodoc]] models.controlnets.controlnet_sparsectrl.SparseControlNetOutput

src/diffusers/models/controlnet.py

@@ -1,115 +0,0 @@
-# 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 Optional, Tuple, Union
-
-from ..utils import deprecate
-from .controlnets.controlnet import (  # noqa
-    ControlNetConditioningEmbedding,
-    ControlNetModel,
-    ControlNetOutput,
-    zero_module,
-)
-
-
-class ControlNetOutput(ControlNetOutput):
-    def __init__(self, *args, **kwargs):
-        deprecation_message = "Importing `ControlNetOutput` from `diffusers.models.controlnet` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet import ControlNetOutput`, instead."
-        deprecate("diffusers.models.controlnet.ControlNetOutput", "0.34", deprecation_message)
-        super().__init__(*args, **kwargs)
-
-
-class ControlNetModel(ControlNetModel):
-    def __init__(
-        self,
-        in_channels: int = 4,
-        conditioning_channels: int = 3,
-        flip_sin_to_cos: bool = True,
-        freq_shift: int = 0,
-        down_block_types: Tuple[str, ...] = (
-            "CrossAttnDownBlock2D",
-            "CrossAttnDownBlock2D",
-            "CrossAttnDownBlock2D",
-            "DownBlock2D",
-        ),
-        mid_block_type: Optional[str] = "UNetMidBlock2DCrossAttn",
-        only_cross_attention: Union[bool, Tuple[bool]] = False,
-        block_out_channels: Tuple[int, ...] = (320, 640, 1280, 1280),
-        layers_per_block: int = 2,
-        downsample_padding: int = 1,
-        mid_block_scale_factor: float = 1,
-        act_fn: str = "silu",
-        norm_num_groups: Optional[int] = 32,
-        norm_eps: float = 1e-5,
-        cross_attention_dim: int = 1280,
-        transformer_layers_per_block: Union[int, Tuple[int, ...]] = 1,
-        encoder_hid_dim: Optional[int] = None,
-        encoder_hid_dim_type: Optional[str] = None,
-        attention_head_dim: Union[int, Tuple[int, ...]] = 8,
-        num_attention_heads: Optional[Union[int, Tuple[int, ...]]] = None,
-        use_linear_projection: bool = False,
-        class_embed_type: Optional[str] = None,
-        addition_embed_type: Optional[str] = None,
-        addition_time_embed_dim: Optional[int] = None,
-        num_class_embeds: Optional[int] = None,
-        upcast_attention: bool = False,
-        resnet_time_scale_shift: str = "default",
-        projection_class_embeddings_input_dim: Optional[int] = None,
-        controlnet_conditioning_channel_order: str = "rgb",
-        conditioning_embedding_out_channels: Optional[Tuple[int, ...]] = (16, 32, 96, 256),
-        global_pool_conditions: bool = False,
-        addition_embed_type_num_heads: int = 64,
-    ):
-        deprecation_message = "Importing `ControlNetModel` from `diffusers.models.controlnet` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet import ControlNetModel`, instead."
-        deprecate("diffusers.models.controlnet.ControlNetModel", "0.34", deprecation_message)
-        super().__init__(
-            in_channels=in_channels,
-            conditioning_channels=conditioning_channels,
-            flip_sin_to_cos=flip_sin_to_cos,
-            freq_shift=freq_shift,
-            down_block_types=down_block_types,
-            mid_block_type=mid_block_type,
-            only_cross_attention=only_cross_attention,
-            block_out_channels=block_out_channels,
-            layers_per_block=layers_per_block,
-            downsample_padding=downsample_padding,
-            mid_block_scale_factor=mid_block_scale_factor,
-            act_fn=act_fn,
-            norm_num_groups=norm_num_groups,
-            norm_eps=norm_eps,
-            cross_attention_dim=cross_attention_dim,
-            transformer_layers_per_block=transformer_layers_per_block,
-            encoder_hid_dim=encoder_hid_dim,
-            encoder_hid_dim_type=encoder_hid_dim_type,
-            attention_head_dim=attention_head_dim,
-            num_attention_heads=num_attention_heads,
-            use_linear_projection=use_linear_projection,
-            class_embed_type=class_embed_type,
-            addition_embed_type=addition_embed_type,
-            addition_time_embed_dim=addition_time_embed_dim,
-            num_class_embeds=num_class_embeds,
-            upcast_attention=upcast_attention,
-            resnet_time_scale_shift=resnet_time_scale_shift,
-            projection_class_embeddings_input_dim=projection_class_embeddings_input_dim,
-            controlnet_conditioning_channel_order=controlnet_conditioning_channel_order,
-            conditioning_embedding_out_channels=conditioning_embedding_out_channels,
-            global_pool_conditions=global_pool_conditions,
-            addition_embed_type_num_heads=addition_embed_type_num_heads,
-        )
-
-
-class ControlNetConditioningEmbedding(ControlNetConditioningEmbedding):
-    def __init__(self, *args, **kwargs):
-        deprecation_message = "Importing `ControlNetConditioningEmbedding` from `diffusers.models.controlnet` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet import ControlNetConditioningEmbedding`, instead."
-        deprecate("diffusers.models.controlnet.ControlNetConditioningEmbedding", "0.34", deprecation_message)
-        super().__init__(*args, **kwargs)

src/diffusers/models/controlnet_flux.py

@@ -1,70 +0,0 @@
-# Copyright 2025 Black Forest Labs, The HuggingFace Team and The InstantX 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 List
-
-from ..utils import deprecate, logging
-from .controlnets.controlnet_flux import FluxControlNetModel, FluxControlNetOutput, FluxMultiControlNetModel
-
-
-logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
-
-
-class FluxControlNetOutput(FluxControlNetOutput):
-    def __init__(self, *args, **kwargs):
-        deprecation_message = "Importing `FluxControlNetOutput` from `diffusers.models.controlnet_flux` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_flux import FluxControlNetOutput`, instead."
-        deprecate("diffusers.models.controlnet_flux.FluxControlNetOutput", "0.34", deprecation_message)
-        super().__init__(*args, **kwargs)
-
-
-class FluxControlNetModel(FluxControlNetModel):
-    def __init__(
-        self,
-        patch_size: int = 1,
-        in_channels: int = 64,
-        num_layers: int = 19,
-        num_single_layers: int = 38,
-        attention_head_dim: int = 128,
-        num_attention_heads: int = 24,
-        joint_attention_dim: int = 4096,
-        pooled_projection_dim: int = 768,
-        guidance_embeds: bool = False,
-        axes_dims_rope: List[int] = [16, 56, 56],
-        num_mode: int = None,
-        conditioning_embedding_channels: int = None,
-    ):
-        deprecation_message = "Importing `FluxControlNetModel` from `diffusers.models.controlnet_flux` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_flux import FluxControlNetModel`, instead."
-        deprecate("diffusers.models.controlnet_flux.FluxControlNetModel", "0.34", deprecation_message)
-        super().__init__(
-            patch_size=patch_size,
-            in_channels=in_channels,
-            num_layers=num_layers,
-            num_single_layers=num_single_layers,
-            attention_head_dim=attention_head_dim,
-            num_attention_heads=num_attention_heads,
-            joint_attention_dim=joint_attention_dim,
-            pooled_projection_dim=pooled_projection_dim,
-            guidance_embeds=guidance_embeds,
-            axes_dims_rope=axes_dims_rope,
-            num_mode=num_mode,
-            conditioning_embedding_channels=conditioning_embedding_channels,
-        )
-
-
-class FluxMultiControlNetModel(FluxMultiControlNetModel):
-    def __init__(self, *args, **kwargs):
-        deprecation_message = "Importing `FluxMultiControlNetModel` from `diffusers.models.controlnet_flux` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_flux import FluxMultiControlNetModel`, instead."
-        deprecate("diffusers.models.controlnet_flux.FluxMultiControlNetModel", "0.34", deprecation_message)
-        super().__init__(*args, **kwargs)

src/diffusers/models/controlnet_sd3.py

@@ -1,68 +0,0 @@
-# Copyright 2025 Stability AI, The HuggingFace Team and The InstantX 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 ..utils import deprecate, logging
-from .controlnets.controlnet_sd3 import SD3ControlNetModel, SD3ControlNetOutput, SD3MultiControlNetModel
-
-
-logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
-
-
-class SD3ControlNetOutput(SD3ControlNetOutput):
-    def __init__(self, *args, **kwargs):
-        deprecation_message = "Importing `SD3ControlNetOutput` from `diffusers.models.controlnet_sd3` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_sd3 import SD3ControlNetOutput`, instead."
-        deprecate("diffusers.models.controlnet_sd3.SD3ControlNetOutput", "0.34", deprecation_message)
-        super().__init__(*args, **kwargs)
-
-
-class SD3ControlNetModel(SD3ControlNetModel):
-    def __init__(
-        self,
-        sample_size: int = 128,
-        patch_size: int = 2,
-        in_channels: int = 16,
-        num_layers: int = 18,
-        attention_head_dim: int = 64,
-        num_attention_heads: int = 18,
-        joint_attention_dim: int = 4096,
-        caption_projection_dim: int = 1152,
-        pooled_projection_dim: int = 2048,
-        out_channels: int = 16,
-        pos_embed_max_size: int = 96,
-        extra_conditioning_channels: int = 0,
-    ):
-        deprecation_message = "Importing `SD3ControlNetModel` from `diffusers.models.controlnet_sd3` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_sd3 import SD3ControlNetModel`, instead."
-        deprecate("diffusers.models.controlnet_sd3.SD3ControlNetModel", "0.34", deprecation_message)
-        super().__init__(
-            sample_size=sample_size,
-            patch_size=patch_size,
-            in_channels=in_channels,
-            num_layers=num_layers,
-            attention_head_dim=attention_head_dim,
-            num_attention_heads=num_attention_heads,
-            joint_attention_dim=joint_attention_dim,
-            caption_projection_dim=caption_projection_dim,
-            pooled_projection_dim=pooled_projection_dim,
-            out_channels=out_channels,
-            pos_embed_max_size=pos_embed_max_size,
-            extra_conditioning_channels=extra_conditioning_channels,
-        )
-
-
-class SD3MultiControlNetModel(SD3MultiControlNetModel):
-    def __init__(self, *args, **kwargs):
-        deprecation_message = "Importing `SD3MultiControlNetModel` from `diffusers.models.controlnet_sd3` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_sd3 import SD3MultiControlNetModel`, instead."
-        deprecate("diffusers.models.controlnet_sd3.SD3MultiControlNetModel", "0.34", deprecation_message)
-        super().__init__(*args, **kwargs)

src/diffusers/models/controlnet_sparsectrl.py

@@ -1,116 +0,0 @@
-# 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 Optional, Tuple, Union
-
-from ..utils import deprecate, logging
-from .controlnets.controlnet_sparsectrl import (  # noqa
-    SparseControlNetConditioningEmbedding,
-    SparseControlNetModel,
-    SparseControlNetOutput,
-    zero_module,
-)
-
-
-logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
-
-
-class SparseControlNetOutput(SparseControlNetOutput):
-    def __init__(self, *args, **kwargs):
-        deprecation_message = "Importing `SparseControlNetOutput` from `diffusers.models.controlnet_sparsectrl` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_sparsectrl import SparseControlNetOutput`, instead."
-        deprecate("diffusers.models.controlnet_sparsectrl.SparseControlNetOutput", "0.34", deprecation_message)
-        super().__init__(*args, **kwargs)
-
-
-class SparseControlNetConditioningEmbedding(SparseControlNetConditioningEmbedding):
-    def __init__(self, *args, **kwargs):
-        deprecation_message = "Importing `SparseControlNetConditioningEmbedding` from `diffusers.models.controlnet_sparsectrl` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_sparsectrl import SparseControlNetConditioningEmbedding`, instead."
-        deprecate(
-            "diffusers.models.controlnet_sparsectrl.SparseControlNetConditioningEmbedding", "0.34", deprecation_message
-        )
-        super().__init__(*args, **kwargs)
-
-
-class SparseControlNetModel(SparseControlNetModel):
-    def __init__(
-        self,
-        in_channels: int = 4,
-        conditioning_channels: int = 4,
-        flip_sin_to_cos: bool = True,
-        freq_shift: int = 0,
-        down_block_types: Tuple[str, ...] = (
-            "CrossAttnDownBlockMotion",
-            "CrossAttnDownBlockMotion",
-            "CrossAttnDownBlockMotion",
-            "DownBlockMotion",
-        ),
-        only_cross_attention: Union[bool, Tuple[bool]] = False,
-        block_out_channels: Tuple[int, ...] = (320, 640, 1280, 1280),
-        layers_per_block: int = 2,
-        downsample_padding: int = 1,
-        mid_block_scale_factor: float = 1,
-        act_fn: str = "silu",
-        norm_num_groups: Optional[int] = 32,
-        norm_eps: float = 1e-5,
-        cross_attention_dim: int = 768,
-        transformer_layers_per_block: Union[int, Tuple[int, ...]] = 1,
-        transformer_layers_per_mid_block: Optional[Union[int, Tuple[int]]] = None,
-        temporal_transformer_layers_per_block: Union[int, Tuple[int, ...]] = 1,
-        attention_head_dim: Union[int, Tuple[int, ...]] = 8,
-        num_attention_heads: Optional[Union[int, Tuple[int, ...]]] = None,
-        use_linear_projection: bool = False,
-        upcast_attention: bool = False,
-        resnet_time_scale_shift: str = "default",
-        conditioning_embedding_out_channels: Optional[Tuple[int, ...]] = (16, 32, 96, 256),
-        global_pool_conditions: bool = False,
-        controlnet_conditioning_channel_order: str = "rgb",
-        motion_max_seq_length: int = 32,
-        motion_num_attention_heads: int = 8,
-        concat_conditioning_mask: bool = True,
-        use_simplified_condition_embedding: bool = True,
-    ):
-        deprecation_message = "Importing `SparseControlNetModel` from `diffusers.models.controlnet_sparsectrl` is deprecated and this will be removed in a future version. Please use `from diffusers.models.controlnets.controlnet_sparsectrl import SparseControlNetModel`, instead."
-        deprecate("diffusers.models.controlnet_sparsectrl.SparseControlNetModel", "0.34", deprecation_message)
-        super().__init__(
-            in_channels=in_channels,
-            conditioning_channels=conditioning_channels,
-            flip_sin_to_cos=flip_sin_to_cos,
-            freq_shift=freq_shift,
-            down_block_types=down_block_types,
-            only_cross_attention=only_cross_attention,
-            block_out_channels=block_out_channels,
-            layers_per_block=layers_per_block,
-            downsample_padding=downsample_padding,
-            mid_block_scale_factor=mid_block_scale_factor,
-            act_fn=act_fn,
-            norm_num_groups=norm_num_groups,
-            norm_eps=norm_eps,
-            cross_attention_dim=cross_attention_dim,
-            transformer_layers_per_block=transformer_layers_per_block,
-            transformer_layers_per_mid_block=transformer_layers_per_mid_block,
-            temporal_transformer_layers_per_block=temporal_transformer_layers_per_block,
-            attention_head_dim=attention_head_dim,
-            num_attention_heads=num_attention_heads,
-            use_linear_projection=use_linear_projection,
-            upcast_attention=upcast_attention,
-            resnet_time_scale_shift=resnet_time_scale_shift,
-            conditioning_embedding_out_channels=conditioning_embedding_out_channels,
-            global_pool_conditions=global_pool_conditions,
-            controlnet_conditioning_channel_order=controlnet_conditioning_channel_order,
-            motion_max_seq_length=motion_max_seq_length,
-            motion_num_attention_heads=motion_num_attention_heads,
-            concat_conditioning_mask=concat_conditioning_mask,
-            use_simplified_condition_embedding=use_simplified_condition_embedding,
-        )