Merge branch 'main' into transformers-v5-pr

docs/source/en/api/pipelines/chroma.md

@@ -99,3 +99,9 @@ image.save("chroma-single-file.png")
 [[autodoc]] ChromaImg2ImgPipeline
 	- all
 	- __call__
+
+## ChromaInpaintPipeline
+
+[[autodoc]] ChromaInpaintPipeline
+  - all
+  - __call__

src/diffusers/__init__.py

@@ -460,6 +460,7 @@ else:
             "BriaFiboPipeline",
             "BriaPipeline",
             "ChromaImg2ImgPipeline",
+            "ChromaInpaintPipeline",
             "ChromaPipeline",
             "ChronoEditPipeline",
             "CLIPImageProjection",
@@ -1186,6 +1187,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             BriaFiboPipeline,
             BriaPipeline,
             ChromaImg2ImgPipeline,
+            ChromaInpaintPipeline,
             ChromaPipeline,
             ChronoEditPipeline,
             CLIPImageProjection,

src/diffusers/models/attention_dispatch.py

@@ -1573,8 +1573,6 @@ def _templated_context_parallel_attention(
     backward_op,
     _parallel_config: Optional["ParallelConfig"] = None,
 ):
-    if attn_mask is not None:
-        raise ValueError("Attention mask is not yet supported for templated attention.")
     if is_causal:
         raise ValueError("Causal attention is not yet supported for templated attention.")
     if enable_gqa:

src/diffusers/models/transformers/transformer_qwenimage.py

@@ -761,11 +761,14 @@ class QwenImageTransformer2DModel(
     _no_split_modules = ["QwenImageTransformerBlock"]
     _skip_layerwise_casting_patterns = ["pos_embed", "norm"]
     _repeated_blocks = ["QwenImageTransformerBlock"]
+    # Make CP plan compatible with https://github.com/huggingface/diffusers/pull/12702
     _cp_plan = {
-        "": {
+        "transformer_blocks.0": {
             "hidden_states": ContextParallelInput(split_dim=1, expected_dims=3, split_output=False),
             "encoder_hidden_states": ContextParallelInput(split_dim=1, expected_dims=3, split_output=False),
-            "encoder_hidden_states_mask": ContextParallelInput(split_dim=1, expected_dims=2, split_output=False),
+        },
+        "transformer_blocks.*": {
+            "modulate_index": ContextParallelInput(split_dim=1, expected_dims=2, split_output=False),
         },
         "pos_embed": {
             0: ContextParallelInput(split_dim=0, expected_dims=2, split_output=True),

src/diffusers/pipelines/__init__.py

@@ -155,7 +155,7 @@ else:
         "AudioLDM2UNet2DConditionModel",
     ]
     _import_structure["blip_diffusion"] = ["BlipDiffusionPipeline"]
-    _import_structure["chroma"] = ["ChromaPipeline", "ChromaImg2ImgPipeline"]
+    _import_structure["chroma"] = ["ChromaPipeline", "ChromaImg2ImgPipeline", "ChromaInpaintPipeline"]
     _import_structure["cogvideo"] = [
         "CogVideoXPipeline",
         "CogVideoXImageToVideoPipeline",
@@ -598,7 +598,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
         from .blip_diffusion import BlipDiffusionPipeline
         from .bria import BriaPipeline
         from .bria_fibo import BriaFiboPipeline
-        from .chroma import ChromaImg2ImgPipeline, ChromaPipeline
+        from .chroma import ChromaImg2ImgPipeline, ChromaInpaintPipeline, ChromaPipeline
         from .chronoedit import ChronoEditPipeline
         from .cogvideo import (
             CogVideoXFunControlPipeline,

src/diffusers/pipelines/chroma/__init__.py

@@ -24,6 +24,7 @@ except OptionalDependencyNotAvailable:
 else:
     _import_structure["pipeline_chroma"] = ["ChromaPipeline"]
     _import_structure["pipeline_chroma_img2img"] = ["ChromaImg2ImgPipeline"]
+    _import_structure["pipeline_chroma_inpainting"] = ["ChromaInpaintPipeline"]
 if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
     try:
         if not (is_transformers_available() and is_torch_available()):
@@ -33,6 +34,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
     else:
         from .pipeline_chroma import ChromaPipeline
         from .pipeline_chroma_img2img import ChromaImg2ImgPipeline
+        from .pipeline_chroma_inpainting import ChromaInpaintPipeline
 else:
     import sys
 

src/diffusers/utils/dummy_torch_and_transformers_objects.py

@@ -632,6 +632,21 @@ class ChromaImg2ImgPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class ChromaInpaintPipeline(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"]