diffusers 0.30.3__py3-none-any.whl → 0.31.0__py3-none-any.whl

diffusers/pipelines/__init__.py

@@ -123,8 +123,16 @@ else:
         "AnimateDiffSDXLPipeline",
         "AnimateDiffSparseControlNetPipeline",
         "AnimateDiffVideoToVideoPipeline",
+        "AnimateDiffVideoToVideoControlNetPipeline",
+    ]
+    _import_structure["flux"] = [
+        "FluxControlNetPipeline",
+        "FluxControlNetImg2ImgPipeline",
+        "FluxControlNetInpaintPipeline",
+        "FluxImg2ImgPipeline",
+        "FluxInpaintPipeline",
+        "FluxPipeline",
     ]
-    _import_structure["flux"] = ["FluxPipeline"]
     _import_structure["audioldm"] = ["AudioLDMPipeline"]
     _import_structure["audioldm2"] = [
         "AudioLDM2Pipeline",
@@ -136,7 +144,9 @@ else:
         "CogVideoXPipeline",
         "CogVideoXImageToVideoPipeline",
         "CogVideoXVideoToVideoPipeline",
+        "CogVideoXFunControlPipeline",
     ]
+    _import_structure["cogview3"] = ["CogView3PlusPipeline"]
     _import_structure["controlnet"].extend(
         [
             "BlipDiffusionControlNetPipeline",
@@ -150,14 +160,17 @@ else:
     )
     _import_structure["pag"].extend(
         [
+            "StableDiffusionControlNetPAGInpaintPipeline",
             "AnimateDiffPAGPipeline",
             "KolorsPAGPipeline",
             "HunyuanDiTPAGPipeline",
             "StableDiffusion3PAGPipeline",
             "StableDiffusionPAGPipeline",
+            "StableDiffusionPAGImg2ImgPipeline",
             "StableDiffusionControlNetPAGPipeline",
             "StableDiffusionXLPAGPipeline",
             "StableDiffusionXLPAGInpaintPipeline",
+            "StableDiffusionXLControlNetPAGImg2ImgPipeline",
             "StableDiffusionXLControlNetPAGPipeline",
             "StableDiffusionXLPAGImg2ImgPipeline",
             "PixArtSigmaPAGPipeline",
@@ -177,6 +190,7 @@ else:
     _import_structure["controlnet_sd3"].extend(
         [
             "StableDiffusion3ControlNetPipeline",
+            "StableDiffusion3ControlNetInpaintingPipeline",
         ]
     )
     _import_structure["deepfloyd_if"] = [
@@ -446,6 +460,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             AnimateDiffPipeline,
             AnimateDiffSDXLPipeline,
             AnimateDiffSparseControlNetPipeline,
+            AnimateDiffVideoToVideoControlNetPipeline,
             AnimateDiffVideoToVideoPipeline,
         )
         from .audioldm import AudioLDMPipeline
@@ -456,7 +471,13 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
         )
         from .aura_flow import AuraFlowPipeline
         from .blip_diffusion import BlipDiffusionPipeline
-        from .cogvideo import CogVideoXImageToVideoPipeline, CogVideoXPipeline, CogVideoXVideoToVideoPipeline
+        from .cogvideo import (
+            CogVideoXFunControlPipeline,
+            CogVideoXImageToVideoPipeline,
+            CogVideoXPipeline,
+            CogVideoXVideoToVideoPipeline,
+        )
+        from .cogview3 import CogView3PlusPipeline
         from .controlnet import (
             BlipDiffusionControlNetPipeline,
             StableDiffusionControlNetImg2ImgPipeline,
@@ -469,9 +490,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
         from .controlnet_hunyuandit import (
             HunyuanDiTControlNetPipeline,
         )
-        from .controlnet_sd3 import (
-            StableDiffusion3ControlNetPipeline,
-        )
+        from .controlnet_sd3 import StableDiffusion3ControlNetInpaintingPipeline, StableDiffusion3ControlNetPipeline
         from .controlnet_xs import (
             StableDiffusionControlNetXSPipeline,
             StableDiffusionXLControlNetXSPipeline,
@@ -498,7 +517,14 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             VersatileDiffusionTextToImagePipeline,
             VQDiffusionPipeline,
         )
-        from .flux import FluxPipeline
+        from .flux import (
+            FluxControlNetImg2ImgPipeline,
+            FluxControlNetInpaintPipeline,
+            FluxControlNetPipeline,
+            FluxImg2ImgPipeline,
+            FluxInpaintPipeline,
+            FluxPipeline,
+        )
         from .hunyuandit import HunyuanDiTPipeline
         from .i2vgen_xl import I2VGenXLPipeline
         from .kandinsky import (
@@ -550,8 +576,11 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             KolorsPAGPipeline,
             PixArtSigmaPAGPipeline,
             StableDiffusion3PAGPipeline,
+            StableDiffusionControlNetPAGInpaintPipeline,
             StableDiffusionControlNetPAGPipeline,
+            StableDiffusionPAGImg2ImgPipeline,
             StableDiffusionPAGPipeline,
+            StableDiffusionXLControlNetPAGImg2ImgPipeline,
             StableDiffusionXLControlNetPAGPipeline,
             StableDiffusionXLPAGImg2ImgPipeline,
             StableDiffusionXLPAGInpaintPipeline,

diffusers/pipelines/animatediff/__init__.py

@@ -26,6 +26,7 @@ else:
     _import_structure["pipeline_animatediff_sdxl"] = ["AnimateDiffSDXLPipeline"]
     _import_structure["pipeline_animatediff_sparsectrl"] = ["AnimateDiffSparseControlNetPipeline"]
     _import_structure["pipeline_animatediff_video2video"] = ["AnimateDiffVideoToVideoPipeline"]
+    _import_structure["pipeline_animatediff_video2video_controlnet"] = ["AnimateDiffVideoToVideoControlNetPipeline"]
 
 if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
     try:
@@ -40,6 +41,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
         from .pipeline_animatediff_sdxl import AnimateDiffSDXLPipeline
         from .pipeline_animatediff_sparsectrl import AnimateDiffSparseControlNetPipeline
         from .pipeline_animatediff_video2video import AnimateDiffVideoToVideoPipeline
+        from .pipeline_animatediff_video2video_controlnet import AnimateDiffVideoToVideoControlNetPipeline
         from .pipeline_output import AnimateDiffPipelineOutput
 
 else:

diffusers/pipelines/animatediff/pipeline_animatediff.py

@@ -432,7 +432,6 @@ class AnimateDiffPipeline(
             extra_step_kwargs["generator"] = generator
         return extra_step_kwargs
 
-    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.check_inputs
     def check_inputs(
         self,
         prompt,
@@ -470,8 +469,8 @@ class AnimateDiffPipeline(
             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)}")
+        elif prompt is not None and not isinstance(prompt, (str, list, dict)):
+            raise ValueError(f"`prompt` has to be of type `str`, `list` or `dict` but is {type(prompt)=}")
 
         if negative_prompt is not None and negative_prompt_embeds is not None:
             raise ValueError(
@@ -557,11 +556,15 @@ class AnimateDiffPipeline(
     def num_timesteps(self):
         return self._num_timesteps
 
+    @property
+    def interrupt(self):
+        return self._interrupt
+
     @torch.no_grad()
     @replace_example_docstring(EXAMPLE_DOC_STRING)
     def __call__(
         self,
-        prompt: Union[str, List[str]] = None,
+        prompt: Optional[Union[str, List[str]]] = None,
         num_frames: Optional[int] = 16,
         height: Optional[int] = None,
         width: Optional[int] = None,
@@ -701,9 +704,10 @@ class AnimateDiffPipeline(
         self._guidance_scale = guidance_scale
         self._clip_skip = clip_skip
         self._cross_attention_kwargs = cross_attention_kwargs
+        self._interrupt = False
 
         # 2. Define call parameters
-        if prompt is not None and isinstance(prompt, str):
+        if prompt is not None and isinstance(prompt, (str, dict)):
             batch_size = 1
         elif prompt is not None and isinstance(prompt, list):
             batch_size = len(prompt)
@@ -716,22 +720,39 @@ class AnimateDiffPipeline(
         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_videos_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 self.free_noise_enabled:
+            prompt_embeds, negative_prompt_embeds = self._encode_prompt_free_noise(
+                prompt=prompt,
+                num_frames=num_frames,
+                device=device,
+                num_videos_per_prompt=num_videos_per_prompt,
+                do_classifier_free_guidance=self.do_classifier_free_guidance,
+                negative_prompt=negative_prompt,
+                prompt_embeds=prompt_embeds,
+                negative_prompt_embeds=negative_prompt_embeds,
+                lora_scale=text_encoder_lora_scale,
+                clip_skip=self.clip_skip,
+            )
+        else:
+            prompt_embeds, negative_prompt_embeds = self.encode_prompt(
+                prompt,
+                device,
+                num_videos_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])
+
+            prompt_embeds = prompt_embeds.repeat_interleave(repeats=num_frames, dim=0)
 
         if ip_adapter_image is not None or ip_adapter_image_embeds is not None:
             image_embeds = self.prepare_ip_adapter_image_embeds(
@@ -783,6 +804,9 @@ class AnimateDiffPipeline(
             # 8. Denoising loop
             with self.progress_bar(total=self._num_timesteps) as progress_bar:
                 for i, t in enumerate(timesteps):
+                    if self.interrupt:
+                        continue
+
                     # 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)

diffusers/pipelines/animatediff/pipeline_animatediff_controlnet.py

@@ -505,8 +505,8 @@ class AnimateDiffControlNetPipeline(
             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)}")
+        elif prompt is not None and not isinstance(prompt, (str, list, dict)):
+            raise ValueError(f"`prompt` has to be of type `str`, `list` or `dict` but is {type(prompt)}")
 
         if negative_prompt is not None and negative_prompt_embeds is not None:
             raise ValueError(
@@ -699,6 +699,10 @@ class AnimateDiffControlNetPipeline(
     def num_timesteps(self):
         return self._num_timesteps
 
+    @property
+    def interrupt(self):
+        return self._interrupt
+
     @torch.no_grad()
     def __call__(
         self,
@@ -858,9 +862,10 @@ class AnimateDiffControlNetPipeline(
         self._guidance_scale = guidance_scale
         self._clip_skip = clip_skip
         self._cross_attention_kwargs = cross_attention_kwargs
+        self._interrupt = False
 
         # 2. Define call parameters
-        if prompt is not None and isinstance(prompt, str):
+        if prompt is not None and isinstance(prompt, (str, dict)):
             batch_size = 1
         elif prompt is not None and isinstance(prompt, list):
             batch_size = len(prompt)
@@ -883,22 +888,39 @@ class AnimateDiffControlNetPipeline(
         text_encoder_lora_scale = (
             cross_attention_kwargs.get("scale", None) if cross_attention_kwargs is not None else None
         )
-        prompt_embeds, negative_prompt_embeds = self.encode_prompt(
-            prompt,
-            device,
-            num_videos_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 self.free_noise_enabled:
+            prompt_embeds, negative_prompt_embeds = self._encode_prompt_free_noise(
+                prompt=prompt,
+                num_frames=num_frames,
+                device=device,
+                num_videos_per_prompt=num_videos_per_prompt,
+                do_classifier_free_guidance=self.do_classifier_free_guidance,
+                negative_prompt=negative_prompt,
+                prompt_embeds=prompt_embeds,
+                negative_prompt_embeds=negative_prompt_embeds,
+                lora_scale=text_encoder_lora_scale,
+                clip_skip=self.clip_skip,
+            )
+        else:
+            prompt_embeds, negative_prompt_embeds = self.encode_prompt(
+                prompt,
+                device,
+                num_videos_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])
+
+            prompt_embeds = prompt_embeds.repeat_interleave(repeats=num_frames, dim=0)
 
         if ip_adapter_image is not None or ip_adapter_image_embeds is not None:
             image_embeds = self.prepare_ip_adapter_image_embeds(
@@ -990,6 +1012,9 @@ class AnimateDiffControlNetPipeline(
             # 8. Denoising loop
             with self.progress_bar(total=self._num_timesteps) as progress_bar:
                 for i, t in enumerate(timesteps):
+                    if self.interrupt:
+                        continue
+
                     # 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)
@@ -1002,7 +1027,6 @@ class AnimateDiffControlNetPipeline(
                     else:
                         control_model_input = latent_model_input
                         controlnet_prompt_embeds = prompt_embeds
-                    controlnet_prompt_embeds = controlnet_prompt_embeds.repeat_interleave(num_frames, dim=0)
 
                     if isinstance(controlnet_keep[i], list):
                         cond_scale = [c * s for c, s in zip(controlnet_conditioning_scale, controlnet_keep[i])]

diffusers/pipelines/animatediff/pipeline_animatediff_sdxl.py

@@ -113,9 +113,21 @@ EXAMPLE_DOC_STRING = """
 
 # 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):
-    """
-    Rescale `noise_cfg` according to `guidance_rescale`. Based on findings of [Common Diffusion Noise Schedules and
-    Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf). See Section 3.4
+    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://arxiv.org/pdf/2305.08891.pdf).
+
+    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)
@@ -135,7 +147,7 @@ def retrieve_timesteps(
     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`.
 
@@ -1143,6 +1155,8 @@ class AnimateDiffSDXLPipeline(
             add_text_embeds = torch.cat([negative_pooled_prompt_embeds, add_text_embeds], dim=0)
             add_time_ids = torch.cat([negative_add_time_ids, add_time_ids], dim=0)
 
+        prompt_embeds = prompt_embeds.repeat_interleave(repeats=num_frames, dim=0)
+
         prompt_embeds = prompt_embeds.to(device)
         add_text_embeds = add_text_embeds.to(device)
         add_time_ids = add_time_ids.to(device).repeat(batch_size * num_videos_per_prompt, 1)

diffusers/pipelines/animatediff/pipeline_animatediff_sparsectrl.py

@@ -878,6 +878,8 @@ class AnimateDiffSparseControlNetPipeline(
         if self.do_classifier_free_guidance:
             prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
 
+        prompt_embeds = prompt_embeds.repeat_interleave(repeats=num_frames, dim=0)
+
         # 4. Prepare IP-Adapter embeddings
         if ip_adapter_image is not None or ip_adapter_image_embeds is not None:
             image_embeds = self.prepare_ip_adapter_image_embeds(