Merge branch 'main' into compile-ci

Author: sayakpaul

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

@@ -52,6 +52,7 @@ The following models are available for the image-to-video pipeline:
 | [`Skywork/SkyReels-V1-Hunyuan-I2V`](https://huggingface.co/Skywork/SkyReels-V1-Hunyuan-I2V) | Skywork's custom finetune of HunyuanVideo (de-distilled). Performs best with `97x544x960` resolution. Performs best at `97x544x960` resolution, `guidance_scale=1.0`, `true_cfg_scale=6.0` and a negative prompt. |
 | [`hunyuanvideo-community/HunyuanVideo-I2V-33ch`](https://huggingface.co/hunyuanvideo-community/HunyuanVideo-I2V) | Tecent's official HunyuanVideo 33-channel I2V model. Performs best at resolutions of 480, 720, 960, 1280. A higher `shift` value when initializing the scheduler is recommended (good values are between 7 and 20). |
 | [`hunyuanvideo-community/HunyuanVideo-I2V`](https://huggingface.co/hunyuanvideo-community/HunyuanVideo-I2V) | Tecent's official HunyuanVideo 16-channel I2V model. Performs best at resolutions of 480, 720, 960, 1280. A higher `shift` value when initializing the scheduler is recommended (good values are between 7 and 20) |
+- [`lllyasviel/FramePackI2V_HY`](https://huggingface.co/lllyasviel/FramePackI2V_HY) | lllyasviel's paper introducing a new technique for long-context video generation called [Framepack](https://arxiv.org/abs/2504.12626). |
 
 ## Quantization
 

src/diffusers/__init__.py

@@ -175,6 +175,7 @@
             "HunyuanDiT2DControlNetModel",
             "HunyuanDiT2DModel",
             "HunyuanDiT2DMultiControlNetModel",
+            "HunyuanVideoFramepackTransformer3DModel",
             "HunyuanVideoTransformer3DModel",
             "I2VGenXLUNet",
             "Kandinsky3UNet",
@@ -376,6 +377,7 @@
             "HunyuanDiTPAGPipeline",
             "HunyuanDiTPipeline",
             "HunyuanSkyreelsImageToVideoPipeline",
+            "HunyuanVideoFramepackPipeline",
             "HunyuanVideoImageToVideoPipeline",
             "HunyuanVideoPipeline",
             "I2VGenXLPipeline",
@@ -770,6 +772,7 @@
             HunyuanDiT2DControlNetModel,
             HunyuanDiT2DModel,
             HunyuanDiT2DMultiControlNetModel,
+            HunyuanVideoFramepackTransformer3DModel,
             HunyuanVideoTransformer3DModel,
             I2VGenXLUNet,
             Kandinsky3UNet,
@@ -950,6 +953,7 @@
             HunyuanDiTPAGPipeline,
             HunyuanDiTPipeline,
             HunyuanSkyreelsImageToVideoPipeline,
+            HunyuanVideoFramepackPipeline,
             HunyuanVideoImageToVideoPipeline,
             HunyuanVideoPipeline,
             I2VGenXLPipeline,

src/diffusers/models/__init__.py

@@ -79,6 +79,7 @@
     _import_structure["transformers.transformer_flux"] = ["FluxTransformer2DModel"]
     _import_structure["transformers.transformer_hidream_image"] = ["HiDreamImageTransformer2DModel"]
     _import_structure["transformers.transformer_hunyuan_video"] = ["HunyuanVideoTransformer3DModel"]
+    _import_structure["transformers.transformer_hunyuan_video_framepack"] = ["HunyuanVideoFramepackTransformer3DModel"]
     _import_structure["transformers.transformer_ltx"] = ["LTXVideoTransformer3DModel"]
     _import_structure["transformers.transformer_lumina2"] = ["Lumina2Transformer2DModel"]
     _import_structure["transformers.transformer_mochi"] = ["MochiTransformer3DModel"]
@@ -156,6 +157,7 @@
             FluxTransformer2DModel,
             HiDreamImageTransformer2DModel,
             HunyuanDiT2DModel,
+            HunyuanVideoFramepackTransformer3DModel,
             HunyuanVideoTransformer3DModel,
             LatteTransformer3DModel,
             LTXVideoTransformer3DModel,

src/diffusers/models/transformers/__init__.py

@@ -23,6 +23,7 @@
     from .transformer_flux import FluxTransformer2DModel
     from .transformer_hidream_image import HiDreamImageTransformer2DModel
     from .transformer_hunyuan_video import HunyuanVideoTransformer3DModel
+    from .transformer_hunyuan_video_framepack import HunyuanVideoFramepackTransformer3DModel
     from .transformer_ltx import LTXVideoTransformer3DModel
     from .transformer_lumina2 import Lumina2Transformer2DModel
     from .transformer_mochi import MochiTransformer3DModel