SkyReels Hunyuan T2V & I2V (huggingface#10837)

* update

* make fix-copies

* update

* tests

* update

* update

* add co-author

Co-Authored-By: Langdx <[email protected]>

* add co-author

Co-Authored-By: howe <[email protected]>

* update

---------

Co-authored-by: Langdx <[email protected]>
Co-authored-by: howe <[email protected]>

Author: 3 people

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

@@ -32,6 +32,21 @@ Recommendations for inference:
 - For smaller resolution videos, try lower values of `shift` (between `2.0` to `5.0`) in the [Scheduler](https://huggingface.co/docs/diffusers/main/en/api/schedulers/flow_match_euler_discrete#diffusers.FlowMatchEulerDiscreteScheduler.shift). For larger resolution images, try higher values (between `7.0` and `12.0`). The default value is `7.0` for HunyuanVideo.
 - For more information about supported resolutions and other details, please refer to the original repository [here](https://github.com/Tencent/HunyuanVideo/).
 
+## Available models
+
+The following models are available for the [`HunyuanVideoPipeline`](text-to-video) pipeline:
+
+| Model name | Description |
+|:---|:---|
+| [`hunyuanvideo-community/HunyuanVideo`](https://huggingface.co/hunyuanvideo-community/HunyuanVideo) | Official HunyuanVideo (guidance-distilled). Performs best at multiple resolutions and frames. Performs best with `guidance_scale=6.0`, `true_cfg_scale=1.0` and without a negative prompt. |
+| [`https://huggingface.co/Skywork/SkyReels-V1-Hunyuan-T2V`](https://huggingface.co/Skywork/SkyReels-V1-Hunyuan-T2V) | Skywork's custom finetune of HunyuanVideo (de-distilled). Performs best with `97x544x960` resolution, `guidance_scale=1.0`, `true_cfg_scale=6.0` and a negative prompt. |
+
+The following models are available for the image-to-video pipeline:
+
+| Model name | Description |
+|:---|:---|
+| [`https://huggingface.co/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. |
+
 ## Quantization
 
 Quantization helps reduce the memory requirements of very large models by storing model weights in a lower precision data type. However, quantization may have varying impact on video quality depending on the video model.

src/diffusers/__init__.py

@@ -305,6 +305,7 @@
             "HunyuanDiTControlNetPipeline",
             "HunyuanDiTPAGPipeline",
             "HunyuanDiTPipeline",
+            "HunyuanSkyreelsImageToVideoPipeline",
             "HunyuanVideoPipeline",
             "I2VGenXLPipeline",
             "IFImg2ImgPipeline",
@@ -804,6 +805,7 @@
             HunyuanDiTControlNetPipeline,
             HunyuanDiTPAGPipeline,
             HunyuanDiTPipeline,
+            HunyuanSkyreelsImageToVideoPipeline,
             HunyuanVideoPipeline,
             I2VGenXLPipeline,
             IFImg2ImgPipeline,

src/diffusers/pipelines/__init__.py

@@ -217,7 +217,7 @@
         "IFSuperResolutionPipeline",
     ]
     _import_structure["hunyuandit"] = ["HunyuanDiTPipeline"]
-    _import_structure["hunyuan_video"] = ["HunyuanVideoPipeline"]
+    _import_structure["hunyuan_video"] = ["HunyuanVideoPipeline", "HunyuanSkyreelsImageToVideoPipeline"]
     _import_structure["kandinsky"] = [
         "KandinskyCombinedPipeline",
         "KandinskyImg2ImgCombinedPipeline",
@@ -558,7 +558,7 @@
             FluxPriorReduxPipeline,
             ReduxImageEncoder,
         )
-        from .hunyuan_video import HunyuanVideoPipeline
+        from .hunyuan_video import HunyuanSkyreelsImageToVideoPipeline, HunyuanVideoPipeline
         from .hunyuandit import HunyuanDiTPipeline
         from .i2vgen_xl import I2VGenXLPipeline
         from .kandinsky import (

src/diffusers/pipelines/hunyuan_video/__init__.py

@@ -22,6 +22,7 @@
 
     _dummy_objects.update(get_objects_from_module(dummy_torch_and_transformers_objects))
 else:
+    _import_structure["pipeline_hunyuan_skyreels_image2video"] = ["HunyuanSkyreelsImageToVideoPipeline"]
     _import_structure["pipeline_hunyuan_video"] = ["HunyuanVideoPipeline"]
 
 if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
@@ -32,6 +33,7 @@
     except OptionalDependencyNotAvailable:
         from ...utils.dummy_torch_and_transformers_objects import *
     else:
+        from .pipeline_hunyuan_skyreels_image2video import HunyuanSkyreelsImageToVideoPipeline
         from .pipeline_hunyuan_video import HunyuanVideoPipeline
 
 else: