1580158
/
automatic
mirror of https://github.com/vladmandic/automatic
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
automatic/pipelines
History
CalamitousFelicitousness 76aa949a26 refactor: integrate sharpfin for high-quality image resize 
Vendor sharpfin library (Apache 2.0) and add centralized wrapper
module (images_sharpfin.py) replacing torchvision tensor/PIL
conversion and resize operations throughout the codebase.

- Add modules/sharpfin/ vendored library with MKS2021, Lanczos3,
  Mitchell, Catmull-Rom kernels and optional Triton sparse acceleration
- Add modules/images_sharpfin.py wrapper with to_tensor(), to_pil(),
  pil_to_tensor(), normalize(), resize(), resize_tensor()
- Add resize_quality and resize_linearize_srgb settings
- Add MKS2021 and Lanczos3 upscaler entries
- Replace torchvision.transforms.functional imports across 18 files
- to_pil() auto-detects HWC/BHWC layout, adds .round() before uint8
- Sparse Triton path falls back to dense GPU on compilation failure
- Mixed-axis resize splits into two single-axis scale() calls
- Masks and non-sRGB data always use linearize=False
 		2026-02-11 09:57:37 +01:00
..
bria cleanup todo 2026-01-19 11:10:05 +01:00
f_lite refactor: integrate sharpfin for high-quality image resize 2026-02-11 09:57:37 +01:00
flex2 Flex.2: fix pipeline bug that forces contol in inpaint 2025-09-03 19:30:57 +06:00
flux refactor: update nunchaku repo URLs and version handling 2026-02-07 22:27:05 +00:00
hdm cleanup todo 2026-01-19 11:10:05 +01:00
hidream major refactoring of modules 2025-07-03 09:18:38 -04:00
meissonic cleanup flux loader 2025-08-11 16:05:39 -04:00
omnigen2 lint formatting 2025-07-05 17:13:31 -04:00
qwen feat: add 4-step Nunchaku variants for Qwen-Lightning models 2026-02-07 22:27:05 +00:00
segmoe major refactoring of modules 2025-07-03 09:18:38 -04:00
wan support configurable multi-stage models in video tab 2025-09-22 13:18:39 -04:00
xomni linting 2025-10-05 20:25:33 -04:00
generic.py refactor: update nunchaku repo URLs and version handling 2026-02-07 22:27:05 +00:00
model_anima.py fix: Anima pipeline detection, custom module loading, and model type 2026-02-02 00:44:51 +00:00
model_auraflow.py fix pony-v7 text-encoder 2025-11-08 08:23:44 -05:00
model_bria.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_chroma.py add chroma-inpaint 2026-01-15 08:37:44 +01:00
model_chrono.py update chrono pipeline 2025-11-10 08:43:50 -05:00
model_cogview.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_cosmos.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_flex.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_flite.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_flux.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_flux2.py Complete Z-Image support 2025-11-27 17:25:36 +03:00
model_flux2_klein.py fix(model_flux2_klein): update generic.py and pipeline 2026-01-17 21:36:44 +00:00
model_glm.py lint and safeguard glm 2026-01-16 09:40:48 +01:00
model_google.py Cleanup 2026-01-10 03:07:44 +00:00
model_hdm.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_hidream.py add nvidia-chronoedit 2025-10-30 19:52:29 -04:00
model_hunyuandit.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_hyimage.py sdnq always-register 2025-11-16 08:14:01 -05:00
model_kandinsky.py add kandinsky5 2025-12-09 09:47:22 +01:00
model_kolors.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_longcat.py fix longcat-edit 2026-01-01 11:22:43 +01:00
model_lumina.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_meissonic.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_nextstep.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_omnigen.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_ovis.py add ovis-image 2025-12-26 08:05:25 +01:00
model_pixart.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_prx.py add prx 2025-11-13 17:10:14 -05:00
model_qwen.py feat: add 4-step Nunchaku variants for Qwen-Lightning models 2026-02-07 22:27:05 +00:00
model_sana.py refactor: update nunchaku repo URLs and version handling 2026-02-07 22:27:05 +00:00
model_sd3.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_stablecascade.py refactor: integrate sharpfin for high-quality image resize 2026-02-11 09:57:37 +01:00
model_wanai.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_xomni.py cleanup dict defined in function args 2025-10-30 03:11:50 +03:00
model_z_image.py refactor: update nunchaku repo URLs and version handling 2026-02-07 22:27:05 +00:00