Add SkyReels V2: Infinite-Length Film Generative Model

[tolgacangoz]

Thanks for the opportunity to fix #11374!

Original repo: https://github.com/SkyworkAI/SkyReels-V2
Paper: https://huggingface.co/papers/2504.13074

SkyReels V2's main contributions are summarized as follow:
• Comprehensive video captioner that understand the shot language while capturing the general description of the video, which dramatically improve the prompt adherence.
• Motion-specific preference optimization enhances motion dynamics with a semi-automatic data collection pipeline.
• Effective Diffusion-forcing adaptation enables the generation of ultra-long videos and story generation capabilities, providing a robust framework for extending temporal coherence and narrative depth.
• SkyCaptioner-V1 and SkyReels-V2 series models including diffusion-forcing, text2video, image2video, camera director and elements2video models with various sizes (1.3B, 5B, 14B) are open-sourced.

TODOs:
⏳ FlowMatchUniPCMultistepScheduler: just copy-pasted from the original repo
✅ SkyReelsV2Transformer3DModel: 90% WanTransformer3DModel
✅ SkyReelsV2DiffusionForcingPipeline

• tolgacangoz/SkyReels-V2-DF-1.3B-540P-Diffusers is ready to be forked.
• tolgacangoz/SkyReels-V2-DF-14B-720P-Diffusers is ready to be forked.
• tolgacangoz/SkyReels-V2-DF-14B-540P-Diffusers is ready to be forked.

⏳ SkyReelsV2DiffusionForcingImageToVideoPipeline: Includes Start/End Frame Control.
⏳ SkyReelsV2DiffusionForcingVideoToVideoPipeline: Extends a given video.
⬜ SkyReelsV2Pipeline
⬜ SkyReelsV2ImageToVideoPipeline
⏳ scripts/convert_skyreelsv2_to_diffusers.py
⬜ Did you make sure to update the documentation with your changes?
⬜ Did you write any new necessary tests?

prompt="A cat and a dog baking a cake together in a kitchen. The cat is carefully measuring
		flour, while the dog is stirring the batter with a wooden spoon. The kitchen is
		cozy, with sunlight streaming through the window."
num_inference_steps=30,
height=544,
width=960,
guidance_scale=6.0,
num_frames=97 or 257,
ar_step=5,
causal_block_size=5,
generator=torch.Generator(device="cuda").manual_seed(0 or 37),
overlap_history=None for short; 17 for long,
addnoise_condition=20

Skywork/SkyReels-V2-DF-1.3B-540P
seed 0 and num_frames 97
Original repo	diffusers integration
original_0_short.mp4	diffusers_0_short.mp4

seed 37 and num_frames 97
Original repo	diffusers integration
original_37_short.mp4	diffusers_37_short.mp4

seed 0 and num_frames 257
Original repo	diffusers integration
original_0_long.mp4	diffusers_0_long.mp4

seed 37 and num_frames 257
Original repo	diffusers integration
original_37_long.mp4	diffusers_37_long.mp4

Firstly, I want to congratulate you on this great work, and thanks for open-sourcing it, SkyReels Team! This PR attempted to integrate your model.
Now, this PR is ready for review for SkyReelsV2Transformer3DModel and SkyReelsV2DiffusionForcingPipeline. Other pipelines will be incoming right after the first feedback...

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

@yiyixuxu @a-r-r-o-w @linoytsaban @yjp999 @Howe2018 @RoseRollZhu @pftq @Langdx @guibinchen @qiudi0127 @nitinmukesh @tin2tin @ukaprch @okaris

[ukaprch]

It's about time. Thanks.

[tolgacangoz]

When we set inference_steps=30 in asynchronous mode (by setting e.g., ar_step=5), the original repository displays 50 steps via its tqdm -step_matrix includes 50 elements; i.e., requires more steps. I integrated this with progress_bar_step = len(timesteps) / len(step_matrix). The user sees 30 steps moving by decimal 0.6 at each step, while there are actually 50. WDYT?

[tolgacangoz]

As a second thought, this might be confusing. Should I arrange with self.progress_bar(total=len(step_matrix)) as progress_bar: so that the user would see the real number of steps directly? And decimal stepping in the progress bar probably isn't something common. The difference between synchronous and asynchronous inferences can be explained in the documentation.