Merge branch 'main' into test-better-torch-compile

Author: sayakpaul

docs/source/en/community_projects.md

@@ -83,4 +83,8 @@ Happy exploring, and thank you for being part of the Diffusers community!
     <td><a href="https://github.com/suzukimain/auto_diffusers"> Model Search </a></td>
     <td>Search models on Civitai and Hugging Face</td>
   </tr>
+  <tr style="border-top: 2px solid black">
+    <td><a href="https://github.com/beinsezii/skrample"> Skrample </a></td>
+    <td>Fully modular scheduler functions with 1st class diffusers integration.</td>
+  </tr>
 </table>

src/diffusers/__init__.py

@@ -860,6 +860,7 @@
             EulerDiscreteScheduler,
             FlowMatchEulerDiscreteScheduler,
             FlowMatchHeunDiscreteScheduler,
+            FlowMatchLCMScheduler,
             HeunDiscreteScheduler,
             IPNDMScheduler,
             KarrasVeScheduler,

src/diffusers/models/transformers/transformer_hidream_image.py

@@ -95,7 +95,12 @@ def forward(self, latent):
 def rope(pos: torch.Tensor, dim: int, theta: int) -> torch.Tensor:
     assert dim % 2 == 0, "The dimension must be even."
 
-    scale = torch.arange(0, dim, 2, dtype=torch.float64, device=pos.device) / dim
+    is_mps = pos.device.type == "mps"
+    is_npu = pos.device.type == "npu"
+
+    dtype = torch.float32 if (is_mps or is_npu) else torch.float64
+
+    scale = torch.arange(0, dim, 2, dtype=dtype, device=pos.device) / dim
     omega = 1.0 / (theta**scale)
 
     batch_size, seq_length = pos.shape

src/diffusers/pipelines/hidream_image/pipeline_hidream_image.py

@@ -36,7 +36,7 @@
         ```py
         >>> import torch
         >>> from transformers import PreTrainedTokenizerFast, LlamaForCausalLM
-        >>> from diffusers import UniPCMultistepScheduler, HiDreamImagePipeline, HiDreamImageTransformer2DModel
+        >>> from diffusers import UniPCMultistepScheduler, HiDreamImagePipeline
 
         >>> scheduler = UniPCMultistepScheduler(
         ...     flow_shift=3.0, prediction_type="flow_prediction", use_flow_sigmas=True
@@ -50,16 +50,11 @@
         ...     torch_dtype=torch.bfloat16,
         ... )
 
-        >>> transformer = HiDreamImageTransformer2DModel.from_pretrained(
-        ...     "HiDream-ai/HiDream-I1-Full", subfolder="transformer", torch_dtype=torch.bfloat16
-        ... )
-
         >>> pipe = HiDreamImagePipeline.from_pretrained(
         ...     "HiDream-ai/HiDream-I1-Full",
         ...     scheduler=scheduler,
         ...     tokenizer_4=tokenizer_4,
         ...     text_encoder_4=text_encoder_4,
-        ...     transformer=transformer,
         ...     torch_dtype=torch.bfloat16,
         ... )
         >>> pipe.enable_model_cpu_offload()

src/diffusers/schedulers/__init__.py

@@ -60,6 +60,7 @@
     _import_structure["scheduling_euler_discrete"] = ["EulerDiscreteScheduler"]
     _import_structure["scheduling_flow_match_euler_discrete"] = ["FlowMatchEulerDiscreteScheduler"]
     _import_structure["scheduling_flow_match_heun_discrete"] = ["FlowMatchHeunDiscreteScheduler"]
+    _import_structure["scheduling_flow_match_lcm"] = ["FlowMatchLCMScheduler"]
     _import_structure["scheduling_heun_discrete"] = ["HeunDiscreteScheduler"]
     _import_structure["scheduling_ipndm"] = ["IPNDMScheduler"]
     _import_structure["scheduling_k_dpm_2_ancestral_discrete"] = ["KDPM2AncestralDiscreteScheduler"]
@@ -161,6 +162,7 @@
         from .scheduling_euler_discrete import EulerDiscreteScheduler
         from .scheduling_flow_match_euler_discrete import FlowMatchEulerDiscreteScheduler
         from .scheduling_flow_match_heun_discrete import FlowMatchHeunDiscreteScheduler
+        from .scheduling_flow_match_lcm import FlowMatchLCMScheduler
         from .scheduling_heun_discrete import HeunDiscreteScheduler
         from .scheduling_ipndm import IPNDMScheduler
         from .scheduling_k_dpm_2_ancestral_discrete import KDPM2AncestralDiscreteScheduler

src/diffusers/utils/dummy_pt_objects.py

@@ -1763,6 +1763,21 @@ def from_pretrained(cls, *args, **kwargs):
         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"]