Skip to content

add ltx2 vae in sana-video; #13229

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yiyixuxu merged 12 commits into from
Mar 18, 2026
Merged

add ltx2 vae in sana-video; #13229

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
cd639f9
add ltx2 vae in sana-video;
lawrence-cj Mar 2, 2026
c03b739
add ltx vae in conversion script;
lawrence-cj Mar 9, 2026
b6a2c97
Merge branch 'main' into sana-video-ltx2vae
lawrence-cj Mar 10, 2026
e114ec1
Merge branch 'main' into sana-video-ltx2vae
lawrence-cj Mar 11, 2026
e3f5bc7
Merge branch 'main' into sana-video-ltx2vae
lawrence-cj Mar 16, 2026
90eb2f8
Update src/diffusers/pipelines/sana_video/pipeline_sana_video.py
lawrence-cj Mar 17, 2026
5c4941b
Update src/diffusers/pipelines/sana_video/pipeline_sana_video.py
lawrence-cj Mar 17, 2026
d1a7852
Merge branch 'main' into sana-video-ltx2vae
lawrence-cj Mar 17, 2026
3f83cf6
Merge branch 'main' into sana-video-ltx2vae
lawrence-cj Mar 17, 2026
fb7d2dc
Merge branch 'main' into sana-video-ltx2vae
lawrence-cj Mar 17, 2026
4fbc382
condition `vae_scale_factor_xxx` related settings on VAE types;
lawrence-cj Mar 16, 2026
bcfc391
make the mean/std depends on vae class;
lawrence-cj Mar 18, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
49 changes: 41 additions & 8 deletions scripts/convert_sana_video_to_diffusers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,7 @@
from transformers import AutoModelForCausalLM, AutoTokenizer

from diffusers import (
AutoencoderKLLTX2Video,
AutoencoderKLWan,
DPMSolverMultistepScheduler,
FlowMatchEulerDiscreteScheduler,
Expand All @@ -24,7 +25,10 @@

CTX = init_empty_weights if is_accelerate_available else nullcontext

ckpt_ids = ["Efficient-Large-Model/SANA-Video_2B_480p/checkpoints/SANA_Video_2B_480p.pth"]
ckpt_ids = [
"Efficient-Large-Model/SANA-Video_2B_480p/checkpoints/SANA_Video_2B_480p.pth",
"Efficient-Large-Model/SANA-Video_2B_720p/checkpoints/SANA_Video_2B_720p_LTXVAE.pth",
]
# https://github.com/NVlabs/Sana/blob/main/inference_video_scripts/inference_sana_video.py


Expand Down Expand Up @@ -92,12 +96,22 @@ def main(args):
if args.video_size == 480:
sample_size = 30 # Wan-VAE: 8xp2 downsample factor
patch_size = (1, 2, 2)
in_channels = 16
out_channels = 16
elif args.video_size == 720:
sample_size = 22 # Wan-VAE: 32xp1 downsample factor
sample_size = 22 # DC-AE-V: 32xp1 downsample factor
patch_size = (1, 1, 1)
in_channels = 32
out_channels = 32
else:
raise ValueError(f"Video size {args.video_size} is not supported.")

if args.vae_type == "ltx2":
sample_size = 22
patch_size = (1, 1, 1)
in_channels = 128
out_channels = 128

for depth in range(layer_num):
# Transformer blocks.
converted_state_dict[f"transformer_blocks.{depth}.scale_shift_table"] = state_dict.pop(
Expand Down Expand Up @@ -182,8 +196,8 @@ def main(args):
# Transformer
with CTX():
transformer_kwargs = {
"in_channels": 16,
"out_channels": 16,
"in_channels": in_channels,
"out_channels": out_channels,
"num_attention_heads": 20,
"attention_head_dim": 112,
"num_layers": 20,
Expand Down Expand Up @@ -235,9 +249,12 @@ def main(args):
else:
print(colored(f"Saving the whole Pipeline containing {args.model_type}", "green", attrs=["bold"]))
# VAE
vae = AutoencoderKLWan.from_pretrained(
"Wan-AI/Wan2.1-T2V-1.3B-Diffusers", subfolder="vae", torch_dtype=torch.float32
)
if args.vae_type == "ltx2":
vae_path = args.vae_path or "Lightricks/LTX-2"
vae = AutoencoderKLLTX2Video.from_pretrained(vae_path, subfolder="vae", torch_dtype=torch.float32)
else:
vae_path = args.vae_path or "Wan-AI/Wan2.1-T2V-1.3B-Diffusers"
vae = AutoencoderKLWan.from_pretrained(vae_path, subfolder="vae", torch_dtype=torch.float32)

# Text Encoder
text_encoder_model_path = "Efficient-Large-Model/gemma-2-2b-it"
Expand Down Expand Up @@ -314,7 +331,23 @@ def main(args):
choices=["flow-dpm_solver", "flow-euler", "uni-pc"],
help="Scheduler type to use.",
)
parser.add_argument("--task", default="t2v", type=str, required=True, help="Task to convert, t2v or i2v.")
parser.add_argument(
"--vae_type",
default="wan",
type=str,
choices=["wan", "ltx2"],
help="VAE type to use for saving full pipeline (ltx2 uses patchify 1x1x1).",
)
parser.add_argument(
"--vae_path",
default=None,
type=str,
required=False,
help="Optional VAE path or repo id. If not set, a default is used per VAE type.",
)
parser.add_argument(
"--task", default="t2v", type=str, required=True, choices=["t2v", "i2v"], help="Task to convert, t2v or i2v."
)
parser.add_argument("--dump_path", default=None, type=str, required=True, help="Path to the output pipeline.")
parser.add_argument("--save_full_pipeline", action="store_true", help="save all the pipeline elements in one.")
parser.add_argument("--dtype", default="fp32", type=str, choices=["fp32", "fp16", "bf16"], help="Weight dtype.")
Expand Down
44 changes: 31 additions & 13 deletions src/diffusers/pipelines/sana_video/pipeline_sana_video.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,7 @@

from ...callbacks import MultiPipelineCallbacks, PipelineCallback
from ...loaders import SanaLoraLoaderMixin
from ...models import AutoencoderDC, AutoencoderKLWan, SanaVideoTransformer3DModel
from ...models import AutoencoderDC, AutoencoderKLLTX2Video, AutoencoderKLWan, SanaVideoTransformer3DModel
from ...schedulers import DPMSolverMultistepScheduler
from ...utils import (
BACKENDS_MAPPING,
Expand Down Expand Up @@ -194,7 +194,7 @@ class SanaVideoPipeline(DiffusionPipeline, SanaLoraLoaderMixin):
The tokenizer used to tokenize the prompt.
text_encoder ([`Gemma2PreTrainedModel`]):
Text encoder model to encode the input prompts.
vae ([`AutoencoderKLWan` or `AutoencoderDCAEV`]):
vae ([`AutoencoderKLWan`, `AutoencoderDC`, or `AutoencoderKLLTX2Video`]):
Variational Auto-Encoder (VAE) Model to encode and decode videos to and from latent representations.
transformer ([`SanaVideoTransformer3DModel`]):
Conditional Transformer to denoise the input latents.
Expand All @@ -213,7 +213,7 @@ def __init__(
self,
tokenizer: GemmaTokenizer | GemmaTokenizerFast,
text_encoder: Gemma2PreTrainedModel,
vae: AutoencoderDC | AutoencoderKLWan,
vae: AutoencoderDC | AutoencoderKLLTX2Video | AutoencoderKLWan,
transformer: SanaVideoTransformer3DModel,
scheduler: DPMSolverMultistepScheduler,
):
Expand All @@ -223,8 +223,19 @@ def __init__(
tokenizer=tokenizer, text_encoder=text_encoder, vae=vae, transformer=transformer, scheduler=scheduler
)

self.vae_scale_factor_temporal = self.vae.config.scale_factor_temporal if getattr(self, "vae", None) else 4
self.vae_scale_factor_spatial = self.vae.config.scale_factor_spatial if getattr(self, "vae", None) else 8
if getattr(self, "vae", None):
if isinstance(self.vae, AutoencoderKLLTX2Video):
self.vae_scale_factor_temporal = self.vae.config.temporal_compression_ratio
self.vae_scale_factor_spatial = self.vae.config.spatial_compression_ratio
elif isinstance(self.vae, (AutoencoderDC, AutoencoderKLWan)):
self.vae_scale_factor_temporal = self.vae.config.scale_factor_temporal
self.vae_scale_factor_spatial = self.vae.config.scale_factor_spatial
else:
self.vae_scale_factor_temporal = 4
self.vae_scale_factor_spatial = 8
else:
self.vae_scale_factor_temporal = 4
self.vae_scale_factor_spatial = 8
Comment on lines +226 to +238
Copy link
Copy Markdown
Member

@sayakpaul sayakpaul Mar 10, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm, should this be conditioned on the class type of the VAE being used?

Copy link
Copy Markdown
Contributor Author

@lawrence-cj lawrence-cj Mar 16, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fix it.


self.vae_scale_factor = self.vae_scale_factor_spatial

Expand Down Expand Up @@ -985,14 +996,21 @@ def __call__(
if is_torch_version(">=", "2.5.0")
else torch_accelerator_module.OutOfMemoryError
)
latents_mean = (
torch.tensor(self.vae.config.latents_mean)
.view(1, self.vae.config.z_dim, 1, 1, 1)
.to(latents.device, latents.dtype)
)
latents_std = 1.0 / torch.tensor(self.vae.config.latents_std).view(1, self.vae.config.z_dim, 1, 1, 1).to(
latents.device, latents.dtype
)
if isinstance(self.vae, AutoencoderKLLTX2Video):
latents_mean = self.vae.latents_mean
latents_std = self.vae.latents_std
z_dim = self.vae.config.latent_channels
elif isinstance(self.vae, AutoencoderKLWan):
latents_mean = torch.tensor(self.vae.config.latents_mean)
latents_std = torch.tensor(self.vae.config.latents_std)
z_dim = self.vae.config.z_dim
else:
latents_mean = torch.zeros(latents.shape[1], device=latents.device, dtype=latents.dtype)
latents_std = torch.ones(latents.shape[1], device=latents.device, dtype=latents.dtype)
z_dim = latents.shape[1]

latents_mean = latents_mean.view(1, z_dim, 1, 1, 1).to(latents.device, latents.dtype)
latents_std = 1.0 / latents_std.view(1, z_dim, 1, 1, 1).to(latents.device, latents.dtype)
latents = latents / latents_std + latents_mean
try:
video = self.vae.decode(latents, return_dict=False)[0]
Expand Down
64 changes: 43 additions & 21 deletions src/diffusers/pipelines/sana_video/pipeline_sana_video_i2v.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@
from ...callbacks import MultiPipelineCallbacks, PipelineCallback
from ...image_processor import PipelineImageInput
from ...loaders import SanaLoraLoaderMixin
from ...models import AutoencoderDC, AutoencoderKLWan, SanaVideoTransformer3DModel
from ...models import AutoencoderDC, AutoencoderKLLTX2Video, AutoencoderKLWan, SanaVideoTransformer3DModel
from ...schedulers import FlowMatchEulerDiscreteScheduler
from ...utils import (
BACKENDS_MAPPING,
Expand Down Expand Up @@ -184,7 +184,7 @@ class SanaImageToVideoPipeline(DiffusionPipeline, SanaLoraLoaderMixin):
The tokenizer used to tokenize the prompt.
text_encoder ([`Gemma2PreTrainedModel`]):
Text encoder model to encode the input prompts.
vae ([`AutoencoderKLWan` or `AutoencoderDCAEV`]):
vae ([`AutoencoderKLWan`, `AutoencoderDC`, or `AutoencoderKLLTX2Video`]):
Variational Auto-Encoder (VAE) Model to encode and decode videos to and from latent representations.
transformer ([`SanaVideoTransformer3DModel`]):
Conditional Transformer to denoise the input latents.
Expand All @@ -203,7 +203,7 @@ def __init__(
self,
tokenizer: GemmaTokenizer | GemmaTokenizerFast,
text_encoder: Gemma2PreTrainedModel,
vae: AutoencoderDC | AutoencoderKLWan,
vae: AutoencoderDC | AutoencoderKLLTX2Video | AutoencoderKLWan,
transformer: SanaVideoTransformer3DModel,
scheduler: FlowMatchEulerDiscreteScheduler,
):
Expand All @@ -213,8 +213,19 @@ def __init__(
tokenizer=tokenizer, text_encoder=text_encoder, vae=vae, transformer=transformer, scheduler=scheduler
)

self.vae_scale_factor_temporal = self.vae.config.scale_factor_temporal if getattr(self, "vae", None) else 4
self.vae_scale_factor_spatial = self.vae.config.scale_factor_spatial if getattr(self, "vae", None) else 8
if getattr(self, "vae", None):
if isinstance(self.vae, AutoencoderKLLTX2Video):
self.vae_scale_factor_temporal = self.vae.config.temporal_compression_ratio
self.vae_scale_factor_spatial = self.vae.config.spatial_compression_ratio
elif isinstance(self.vae, (AutoencoderDC, AutoencoderKLWan)):
self.vae_scale_factor_temporal = self.vae.config.scale_factor_temporal
self.vae_scale_factor_spatial = self.vae.config.scale_factor_spatial
else:
self.vae_scale_factor_temporal = 4
self.vae_scale_factor_spatial = 8
else:
self.vae_scale_factor_temporal = 4
self.vae_scale_factor_spatial = 8

self.vae_scale_factor = self.vae_scale_factor_spatial

Expand Down Expand Up @@ -687,14 +698,18 @@ def prepare_latents(
image_latents = retrieve_latents(self.vae.encode(image), sample_mode="argmax")
image_latents = image_latents.repeat(batch_size, 1, 1, 1, 1)

latents_mean = (
torch.tensor(self.vae.config.latents_mean)
.view(1, -1, 1, 1, 1)
.to(image_latents.device, image_latents.dtype)
)
latents_std = 1.0 / torch.tensor(self.vae.config.latents_std).view(1, -1, 1, 1, 1).to(
image_latents.device, image_latents.dtype
)
if isinstance(self.vae, AutoencoderKLLTX2Video):
_latents_mean = self.vae.latents_mean
_latents_std = self.vae.latents_std
elif isinstance(self.vae, AutoencoderKLWan):
_latents_mean = torch.tensor(self.vae.config.latents_mean)
_latents_std = torch.tensor(self.vae.config.latents_std)
else:
_latents_mean = torch.zeros(image_latents.shape[1], device=image_latents.device, dtype=image_latents.dtype)
_latents_std = torch.ones(image_latents.shape[1], device=image_latents.device, dtype=image_latents.dtype)

latents_mean = _latents_mean.view(1, -1, 1, 1, 1).to(image_latents.device, image_latents.dtype)
latents_std = 1.0 / _latents_std.view(1, -1, 1, 1, 1).to(image_latents.device, image_latents.dtype)
image_latents = (image_latents - latents_mean) * latents_std

latents[:, :, 0:1] = image_latents.to(dtype)
Expand Down Expand Up @@ -1034,14 +1049,21 @@ def __call__(
if is_torch_version(">=", "2.5.0")
else torch_accelerator_module.OutOfMemoryError
)
latents_mean = (
torch.tensor(self.vae.config.latents_mean)
.view(1, self.vae.config.z_dim, 1, 1, 1)
.to(latents.device, latents.dtype)
)
latents_std = 1.0 / torch.tensor(self.vae.config.latents_std).view(1, self.vae.config.z_dim, 1, 1, 1).to(
latents.device, latents.dtype
)
if isinstance(self.vae, AutoencoderKLLTX2Video):
latents_mean = self.vae.latents_mean
latents_std = self.vae.latents_std
z_dim = self.vae.config.latent_channels
elif isinstance(self.vae, AutoencoderKLWan):
latents_mean = torch.tensor(self.vae.config.latents_mean)
latents_std = torch.tensor(self.vae.config.latents_std)
z_dim = self.vae.config.z_dim
else:
latents_mean = torch.zeros(latents.shape[1], device=latents.device, dtype=latents.dtype)
latents_std = torch.ones(latents.shape[1], device=latents.device, dtype=latents.dtype)
z_dim = latents.shape[1]

latents_mean = latents_mean.view(1, z_dim, 1, 1, 1).to(latents.device, latents.dtype)
latents_std = 1.0 / latents_std.view(1, z_dim, 1, 1, 1).to(latents.device, latents.dtype)
latents = latents / latents_std + latents_mean
try:
video = self.vae.decode(latents, return_dict=False)[0]
Expand Down
Loading