AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-27 17:22:53 +03:00
Code Activity

Add transformer cache context for SkyReels-V2 pipelines & Update docs (#12837)

Browse Source 
* feat: Add transformer cache context for conditional and unconditional predictions for skyreels-v2 pipes.

* docs: Remove SkyReels-V2 FLF2V model link and add contributor attribution.
This commit is contained in:
Tolga Cangöz
2026-01-07 11:30:30 +03:00
committed by GitHub
parent b4be29bda2
commit 8f30bfff1f
6 changed files with 74 additions and 60 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
docs/source/en/api/pipelines/skyreels_v2.md
View File

@@ -37,7 +37,8 @@ The following SkyReels-V2 models are supported in Diffusers:
- [SkyReels-V2 I2V 1.3B - 540P](https://huggingface.co/Skywork/SkyReels-V2-I2V-1.3B-540P-Diffusers)
- [SkyReels-V2 I2V 14B - 540P](https://huggingface.co/Skywork/SkyReels-V2-I2V-14B-540P-Diffusers)
- [SkyReels-V2 I2V 14B - 720P](https://huggingface.co/Skywork/SkyReels-V2-I2V-14B-720P-Diffusers)
- [SkyReels-V2 FLF2V 1.3B - 540P](https://huggingface.co/Skywork/SkyReels-V2-FLF2V-1.3B-540P-Diffusers)
This model was contributed by [M. Tolga Cangöz](https://github.com/tolgacangoz).
> [!TIP]
> Click on the SkyReels-V2 models in the right sidebar for more examples of video generation.

24
src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2.py
View File

@@ -545,22 +545,24 @@ class SkyReelsV2Pipeline(DiffusionPipeline, SkyReelsV2LoraLoaderMixin):
latent_model_input = latents.to(transformer_dtype)
timestep = t.expand(latents.shape[0])
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=prompt_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
noise_uncond = self.transformer(
with self.transformer.cache_context("cond"):
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
encoder_hidden_states=prompt_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
with self.transformer.cache_context("uncond"):
noise_uncond = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
noise_pred = noise_uncond + guidance_scale * (noise_pred - noise_uncond)
# compute the previous noisy sample x_t -> x_t-1

27
src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing.py
View File

@@ -887,25 +887,28 @@ class SkyReelsV2DiffusionForcingPipeline(DiffusionPipeline, SkyReelsV2LoraLoader
)
timestep[:, valid_interval_start:prefix_video_latents_frames] = addnoise_condition
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
noise_uncond = self.transformer(
with self.transformer.cache_context("cond"):
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
encoder_hidden_states=prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
with self.transformer.cache_context("uncond"):
noise_uncond = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
noise_pred = noise_uncond + guidance_scale * (noise_pred - noise_uncond)
update_mask_i = step_update_mask[i]

27
src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing_i2v.py
View File

@@ -966,25 +966,28 @@ class SkyReelsV2DiffusionForcingImageToVideoPipeline(DiffusionPipeline, SkyReels
)
timestep[:, valid_interval_start:prefix_video_latents_frames] = addnoise_condition
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
noise_uncond = self.transformer(
with self.transformer.cache_context("cond"):
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
encoder_hidden_states=prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
with self.transformer.cache_context("uncond"):
noise_uncond = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
noise_pred = noise_uncond + guidance_scale * (noise_pred - noise_uncond)
update_mask_i = step_update_mask[i]

27
src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing_v2v.py
View File

@@ -974,25 +974,28 @@ class SkyReelsV2DiffusionForcingVideoToVideoPipeline(DiffusionPipeline, SkyReels
)
timestep[:, valid_interval_start:prefix_video_latents_frames] = addnoise_condition
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
noise_uncond = self.transformer(
with self.transformer.cache_context("cond"):
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
encoder_hidden_states=prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
with self.transformer.cache_context("uncond"):
noise_uncond = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
enable_diffusion_forcing=True,
fps=fps_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
noise_pred = noise_uncond + guidance_scale * (noise_pred - noise_uncond)
update_mask_i = step_update_mask[i]

26
src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_i2v.py
View File

@@ -678,24 +678,26 @@ class SkyReelsV2ImageToVideoPipeline(DiffusionPipeline, SkyReelsV2LoraLoaderMixi
latent_model_input = torch.cat([latents, condition], dim=1).to(transformer_dtype)
timestep = t.expand(latents.shape[0])
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=prompt_embeds,
encoder_hidden_states_image=image_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
noise_uncond = self.transformer(
with self.transformer.cache_context("cond"):
noise_pred = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
encoder_hidden_states=prompt_embeds,
encoder_hidden_states_image=image_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
if self.do_classifier_free_guidance:
with self.transformer.cache_context("uncond"):
noise_uncond = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
encoder_hidden_states=negative_prompt_embeds,
encoder_hidden_states_image=image_embeds,
attention_kwargs=attention_kwargs,
return_dict=False,
)[0]
noise_pred = noise_uncond + guidance_scale * (noise_pred - noise_uncond)
# compute the previous noisy sample x_t -> x_t-1