Remove unused index_type arg to preprocess_conditions

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

src/diffusers/__init__.py

@@ -569,6 +569,7 @@ else:
             "LEditsPPPipelineStableDiffusionXL",
             "LongCatImageEditPipeline",
             "LongCatImagePipeline",
+            "LTX2ConditionPipeline",
             "LTX2ImageToVideoPipeline",
             "LTX2LatentUpsamplePipeline",
             "LTX2Pipeline",
@@ -1324,6 +1325,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             LEditsPPPipelineStableDiffusionXL,
             LongCatImageEditPipeline,
             LongCatImagePipeline,
+            LTX2ConditionPipeline,
             LTX2ImageToVideoPipeline,
             LTX2LatentUpsamplePipeline,
             LTX2Pipeline,

src/diffusers/pipelines/__init__.py

@@ -292,7 +292,12 @@ else:
         "LTXLatentUpsamplePipeline",
         "LTXI2VLongMultiPromptPipeline",
     ]
-    _import_structure["ltx2"] = ["LTX2Pipeline", "LTX2ImageToVideoPipeline", "LTX2LatentUpsamplePipeline"]
+    _import_structure["ltx2"] = [
+        "LTX2Pipeline",
+        "LTX2ConditionPipeline",
+        "LTX2ImageToVideoPipeline",
+        "LTX2LatentUpsamplePipeline",
+    ]
     _import_structure["lumina"] = ["LuminaPipeline", "LuminaText2ImgPipeline"]
     _import_structure["lumina2"] = ["Lumina2Pipeline", "Lumina2Text2ImgPipeline"]
     _import_structure["lucy"] = ["LucyEditPipeline"]
@@ -745,7 +750,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             LTXLatentUpsamplePipeline,
             LTXPipeline,
         )
-        from .ltx2 import LTX2ImageToVideoPipeline, LTX2LatentUpsamplePipeline, LTX2Pipeline
+        from .ltx2 import LTX2ConditionPipeline, LTX2ImageToVideoPipeline, LTX2LatentUpsamplePipeline, LTX2Pipeline
         from .lucy import LucyEditPipeline
         from .lumina import LuminaPipeline, LuminaText2ImgPipeline
         from .lumina2 import Lumina2Pipeline, Lumina2Text2ImgPipeline

src/diffusers/pipelines/ltx2/__init__.py

@@ -25,6 +25,7 @@ else:
     _import_structure["connectors"] = ["LTX2TextConnectors"]
     _import_structure["latent_upsampler"] = ["LTX2LatentUpsamplerModel"]
     _import_structure["pipeline_ltx2"] = ["LTX2Pipeline"]
+    _import_structure["pipeline_ltx2_condition"] = ["LTX2ConditionPipeline"]
     _import_structure["pipeline_ltx2_image2video"] = ["LTX2ImageToVideoPipeline"]
     _import_structure["pipeline_ltx2_latent_upsample"] = ["LTX2LatentUpsamplePipeline"]
     _import_structure["vocoder"] = ["LTX2Vocoder"]
@@ -40,6 +41,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
         from .connectors import LTX2TextConnectors
         from .latent_upsampler import LTX2LatentUpsamplerModel
         from .pipeline_ltx2 import LTX2Pipeline
+        from .pipeline_ltx2_condition import LTX2ConditionPipeline
         from .pipeline_ltx2_image2video import LTX2ImageToVideoPipeline
         from .pipeline_ltx2_latent_upsample import LTX2LatentUpsamplePipeline
         from .vocoder import LTX2Vocoder

src/diffusers/utils/dummy_torch_and_transformers_objects.py

@@ -2117,6 +2117,21 @@ class LongCatImagePipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class LTX2ConditionPipeline(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 LTX2ImageToVideoPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 

src/diffusers/video_processor.py

@@ -25,9 +25,9 @@ from .image_processor import VaeImageProcessor, is_valid_image, is_valid_image_i
 class VideoProcessor(VaeImageProcessor):
     r"""Simple video processor."""
 
-    def preprocess_video(self, video, height: int | None = None, width: int | None = None) -> torch.Tensor:
+    def preprocess_video(self, video, height: int | None = None, width: int | None = None, **kwargs) -> torch.Tensor:
         r"""
-        Preprocesses input video(s).
+        Preprocesses input video(s). Keyword arguments will be forwarded to `VaeImageProcessor.preprocess`.
 
         Args:
             video (`list[PIL.Image]`, `list[list[PIL.Image]]`, `torch.Tensor`, `np.array`, `list[torch.Tensor]`, `list[np.array]`):
@@ -49,6 +49,10 @@ class VideoProcessor(VaeImageProcessor):
             width (`int`, *optional*`, defaults to `None`):
                 The width in preprocessed frames of the video. If `None`, will use get_default_height_width()` to get
                 the default width.
+
+        Returns:
+            `torch.Tensor` of shape `(batch_size, num_channels, num_frames, height, width)`:
+                A 5D tensor holding the batched channels-first video(s).
         """
         if isinstance(video, list) and isinstance(video[0], np.ndarray) and video[0].ndim == 5:
             warnings.warn(
@@ -79,7 +83,7 @@ class VideoProcessor(VaeImageProcessor):
                 "Input is in incorrect format. Currently, we only support numpy.ndarray, torch.Tensor, PIL.Image.Image"
             )
 
-        video = torch.stack([self.preprocess(img, height=height, width=width) for img in video], dim=0)
+        video = torch.stack([self.preprocess(img, height=height, width=width, **kwargs) for img in video], dim=0)
 
         # move the number of channels before the number of frames.
         video = video.permute(0, 2, 1, 3, 4)
@@ -87,10 +91,11 @@ class VideoProcessor(VaeImageProcessor):
         return video
 
     def postprocess_video(
-        self, video: torch.Tensor, output_type: str = "np"
+        self, video: torch.Tensor, output_type: str = "np", **kwargs
     ) -> np.ndarray | torch.Tensor | list[PIL.Image.Image]:
         r"""
-        Converts a video tensor to a list of frames for export.
+        Converts a video tensor to a list of frames for export. Keyword arguments will be forwarded to
+        `VaeImageProcessor.postprocess`.
 
         Args:
             video (`torch.Tensor`): The video as a tensor.
@@ -100,7 +105,7 @@ class VideoProcessor(VaeImageProcessor):
         outputs = []
         for batch_idx in range(batch_size):
             batch_vid = video[batch_idx].permute(1, 0, 2, 3)
-            batch_output = self.postprocess(batch_vid, output_type)
+            batch_output = self.postprocess(batch_vid, output_type, **kwargs)
             outputs.append(batch_output)
 
         if output_type == "np":