rattus 4e6a1b66a9 speed up and reduce VRAM of QWEN VAE and WAN (less so) (#12036) ...

* ops: introduce autopad for conv3d

This works around pytorch missing ability to causal pad as part of the
kernel and avoids massive weight duplications for padding.

* wan-vae: rework causal padding

This currently uses F.pad which takes a full deep copy and is liable to
be the VRAM peak. Instead, kick spatial padding back to the op and
consolidate the temporal padding with the cat for the cache.

* wan-vae: implement zero pad fast path

The WAN VAE is also QWEN where it is used single-image. These
convolutions are however zero padded 3d convolutions, which means the
VAE is actually just 2D down the last element of the conv weight in
the temporal dimension. Fast path this, to avoid adding zeros that
then just evaporate in convoluton math but cost computation.

2026-01-23 19:56:14 -05:00

..

model_animate.py

[BlockInfo] Wan (#10845)

2025-12-15 17:59:16 -08:00

model_multitalk.py

Support Multi/InfiniteTalk (#10179)

2026-01-21 23:09:48 -05:00

model.py

Support Multi/InfiniteTalk (#10179)

2026-01-21 23:09:48 -05:00

vae2_2.py

WAN2.2: Fix cache VRAM leak on error (#10308)

2025-10-13 15:23:11 -04:00

vae.py

speed up and reduce VRAM of QWEN VAE and WAN (less so) (#12036)

2026-01-23 19:56:14 -05:00