Hongsheng Liu dd6f32cb47 [Perf][Diffusion] Add LTX2 channels-last VAE path ...

Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>

2026-08-11 07:38:08 +08:00

..

__init__.py

[FEATURE] Tile/Patch parallelism refactor for easily support other models (#1366)

2026-03-03 15:48:51 +08:00

autoencoder_kl_flux2.py

[Diffusion] Add VAE patch parallel support for FLUX.2-dev (#5292)

2026-07-24 07:00:57 +00:00

autoencoder_kl_hunyuan_video_15.py

[Diffusion] Support USP and VAE patch parallel for HunyuanVideo 1.5 (#3979)

2026-05-30 14:40:42 +08:00

autoencoder_kl_hunyuan.py

[Feat] Enable VAE parallel in HunyuanImage3 (#3091)

2026-05-20 09:55:33 +08:00

autoencoder_kl_ltx2.py

[Perf][Diffusion] Add LTX2 channels-last VAE path

2026-08-11 07:38:08 +08:00

autoencoder_kl_qwenimage.py

[Feature] Support vae tiling parallel encode (#2368)

2026-04-06 14:47:45 +00:00

autoencoder_kl_wan.py

[Feature] Spatially-sharded (SP) decode for the Wan VAE (#4620)

2026-06-27 09:01:25 +08:00

autoencoder_kl.py

[Quantization] Redo Z-Image text encoder FP8 online quantization (#3279)

2026-05-06 16:37:36 +00:00

distributed_vae_executor.py

[Feature] Spatially-sharded (SP) decode for the Wan VAE (#4620)

2026-06-27 09:01:25 +08:00

wan_spatial_shard.py

[Feature] Spatially-sharded (SP) decode for the Wan VAE (#4620)

2026-06-27 09:01:25 +08:00