CalamitousFelicitousness 1f232f43ef fix(video): keep the conditioner vision tower unquantized ...

The modular text encoder quant config now excludes the vision tower.
The default skip pool covers diffusion module names, so qwen-vl vision
blocks quantized alongside the language layers; vision blocks have no
validated quantization precedent and run only for keyframe workflows.

2026-08-10 22:11:57 +01:00

..

__init__.py

linting

2026-07-03 22:05:33 +02:00

google_veo.py

linting

2026-07-03 22:05:33 +02:00

models_def.py

fix(model): list base-registered modular pipelines in the dropdown

2026-08-10 22:11:57 +01:00

video_cache.py

cleanup logger

2026-02-19 11:09:13 +01:00

video_load.py

feat(video): add minimax h3 joint video and audio model

2026-08-10 22:11:57 +01:00

video_modular.py

fix(video): keep the conditioner vision tower unquantized

2026-08-10 22:11:57 +01:00

video_overrides.py

feat(video): add minimax h3 joint video and audio model

2026-08-10 22:11:57 +01:00

video_run.py

feat(video): add minimax h3 joint video and audio model

2026-08-10 22:11:57 +01:00

video_save.py

add prompt enhance to ltx and fix audio

2026-07-28 13:11:21 +02:00

video_ui.py

feat(video): add minimax h3 joint video and audio model

2026-08-10 22:11:57 +01:00

video_utils.py

feat(video): support first-last-frame for Wan 2.2 I2V

2026-06-30 23:00:36 +01:00

video_vae.py

linting

2026-07-03 22:05:33 +02:00