Skip to content

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.

Support Pruna LTX VAE (#15129)

Jukka Seppänenc38171ddb933
diff --git a/comfy/ldm/lightricks/vae/causal_conv3d.py b/comfy/ldm/lightricks/vae/causal_conv3d.py
index 7515f0d4e..bb1803f12 100644
--- a/comfy/ldm/lightricks/vae/causal_conv3d.py
+++ b/comfy/ldm/lightricks/vae/causal_conv3d.py
@@ -49,6 +49,12 @@ class CausalConv3d(nn.Module):
)
self.temporal_cache_state={}
+ def _empty_output(self, x):
+ # empty (0 frame) outputs must still have the conv's output channels and spatial dims
+ h = (x.shape[3] + 2 * self.conv.padding[1] - self.conv.kernel_size[1]) // self.conv.stride[1] + 1
+ w = (x.shape[4] + 2 * self.conv.padding[2] - self.conv.kernel_size[2]) // self.conv.stride[2] + 1
+ return x.new_empty((x.shape[0], self.out_channels, 0, h, w))
+
def forward(self, x, causal: bool = True):
tid = threading.get_ident()
@@ -58,7 +64,7 @@ class CausalConv3d(nn.Module):
if not causal:
padding_length = padding_length // 2
if x.shape[2] == 0:
- return x
+ return self._empty_output(x)
cached = x[:, :, :1, :, :].repeat((1, 1, padding_length, 1, 1))
pieces = [ cached, x ]
if is_end and not causal:
@@ -83,7 +89,7 @@ class CausalConv3d(nn.Module):
elif is_end:
self.temporal_cache_state[tid] = (None, True)
- return self.conv(x) if x.shape[2] >= self.time_kernel_size else x[:, :, :0, :, :]
+ return self.conv(x) if x.shape[2] >= self.time_kernel_size else self._empty_output(x)
@property
def weight(self):
diff --git a/comfy/ldm/lightricks/vae/causal_video_autoencoder.py b/comfy/ldm/lightricks/vae/causal_video_autoencoder.py
index 5975015e2..5d0eec5b8 100644
--- a/comfy/ldm/lightricks/vae/causal_video_autoencoder.py
+++ b/comfy/ldm/lightricks/vae/causal_video_autoencoder.py
@@ -390,10 +390,10 @@ class Decoder(nn.Module):
# Compute output channel to be product of all channel-multiplier blocks
output_channel = base_channels
- for block_name, block_params in list(reversed(blocks)):
+ for block_name, block_params in blocks:
block_params = block_params if isinstance(block_params, dict) else {}
if block_name == "res_x_y":
- output_channel = output_channel * block_params.get("multiplier", 2)
+ output_channel = block_params.get("in_channels", output_channel * block_params.get("multiplier", 2))
if block_name == "compress_all":
output_channel = output_channel * block_params.get("multiplier", 1)
if block_name == "compress_space":
@@ -432,7 +432,7 @@ class Decoder(nn.Module):
spatial_padding_mode=spatial_padding_mode,
)
elif block_name == "res_x_y":
- output_channel = output_channel // block_params.get("multiplier", 2)
+ output_channel = block_params.get("out_channels", output_channel // block_params.get("multiplier", 2))
block = ResnetBlock3D(
dims=dims,
in_channels=input_channel,