CalamitousFelicitousness 6ac4c0e1a9 fix(krea2): resolve default guidance per variant instead of static 1.0 ...

cfg=-1 means use the pipeline's own default. Krea2Pipeline serves both Base
and Turbo under one class, so a static signature default cannot fit both: Base
needs real guidance while the distilled Turbo runs guidance-free. Resolve the
unset default from is_distilled (Base 4.5, Turbo 1.0). Explicit cfg is passed
through unchanged.

2026-07-06 01:03:09 +01:00

..

anima

update hf hijack, simplify anima loader, add few more models

2026-07-01 22:47:07 +02:00

boogu

add boogu

2026-07-04 15:26:36 +02:00

bria

full codespell coverage

2026-06-04 12:36:10 +02:00

chroma

refactor(lora): centralize universal passthrough prefixes in the shared resolver

2026-06-14 18:22:41 +01:00

chrono

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

cogview

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

cosmos

update transformers/diffusers and enhanced tests

2026-06-12 15:15:16 +02:00

ernie

refactor(lora): centralize universal passthrough prefixes in the shared resolver

2026-06-14 18:22:41 +01:00

f_lite

minor fixes from audit

2026-07-01 11:08:04 +02:00

flex2

PEP 484 pipelines directory

2026-03-25 16:56:41 -07:00

flux

refactor(lora): centralize universal passthrough prefixes in the shared resolver

2026-06-14 18:22:41 +01:00

flux2_klein

refactor(native-loader): replace required_markers with reactive fallback

2026-06-04 21:20:10 +01:00

glm

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

hidream

full codespell coverage

2026-06-04 12:36:10 +02:00

hunyuandit

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

hyimage

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

ideogram

add ideogram prequant and set default schedule

2026-06-08 17:20:12 +02:00

joy

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

kandinsky3

refactor(pipelines): split kandinsky into kandinsky3 and kandinsky5

2026-05-31 00:24:33 +01:00

kandinsky5

refactor(pipelines): split kandinsky into kandinsky3 and kandinsky5

2026-05-31 00:24:33 +01:00

krea2

fix(krea2): resolve default guidance per variant instead of static 1.0

2026-07-06 01:03:09 +01:00

lens

feat(pipelines): wire native_spec for Qwen-Image and custom-class arches

2026-05-31 00:24:33 +01:00

longcat

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

lumina_dimmo

schedulers remove false positives

2026-06-08 09:14:48 +02:00

meissonic

minor fixes from audit

2026-07-01 11:08:04 +02:00

nucleus

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

omnigen2

strict ruff without exclude

2026-05-11 08:36:49 +02:00

ovis

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

pixart

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

prx

refactor(pipelines): move per-arch transformer specs into packages

2026-05-31 00:24:33 +01:00

qwen

docs: cleanup

2026-05-31 00:24:33 +01:00

segmoe

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

step1x

feat(pipelines): wire native_spec for Qwen-Image and custom-class arches

2026-05-31 00:24:33 +01:00

ultraflux

full codespell coverage

2026-06-04 12:36:10 +02:00

vibe

full codespell coverage

2026-06-04 12:36:10 +02:00

wan

PEP 484 pipelines directory

2026-03-25 16:56:41 -07:00

xomni

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

z_image

refactor(lora): centralize universal passthrough prefixes in the shared resolver

2026-06-14 18:22:41 +01:00

zetachroma

ai-agents

2026-04-12 18:54:22 +02:00

generic_map.py

attempt to load incomplete model

2026-06-29 13:39:46 +02:00

generic_shared.py

add boogu

2026-07-04 15:26:36 +02:00

generic_test.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

generic_text_encoder.py

fix(te): load non-T5 single-file text-encoder overrides as their actual class

2026-07-05 21:48:59 +01:00

generic_transformer.py

force safetensors

2026-07-01 14:09:08 +02:00

generic_util.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

generic_vae.py

fix: mutable default arguments + wrong cfg var gating eva_clip text checkpoint download

2026-06-12 05:08:03 -04:00

generic.py

attempt to load incomplete model

2026-06-29 13:39:46 +02:00

model_anima.py

update hf hijack, simplify anima loader, add few more models

2026-07-01 22:47:07 +02:00

model_auraflow.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_boogu.py

add boogu

2026-07-04 15:26:36 +02:00

model_bria.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

model_chroma.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_chrono.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_cogview.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_cosmos.py

update transformers/diffusers and enhanced tests

2026-06-12 15:15:16 +02:00

model_ernie.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_flex.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

model_flite.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

model_flux2_klein.py

add experimental openai interface

2026-07-02 14:36:44 +02:00

model_flux2.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_flux.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_glm.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_google.py

Nano Banana lite add

2026-07-01 20:51:06 +00:00

model_hidream.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

model_hunyuandit.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_hyimage.py

fix: dead hasattr-on-dict path, double video-vae normalization, undefined self.device, hyimage if/elif, attention-mask attr typo, wrong face logged

2026-06-12 13:00:57 -04:00

model_ideogram4.py

logical-vs-physical param count

2026-06-08 12:03:37 +02:00

model_instaflow.py

minor fixes from audit

2026-07-01 11:08:04 +02:00

model_joy.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_kandinsky.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_kolors.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_krea2.py

feat(krea2): add Krea 2 (K2) image model support

2026-06-23 04:41:54 +01:00

model_lens.py

update lens and bria

2026-06-29 09:19:29 +02:00

model_longcat.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_lumina.py

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

model_meissonic.py

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

model_nextstep.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_nucleus.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_omnigen.py

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

model_ovis.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_pixart.py

fixes based on full skills run

2026-06-04 11:10:26 +02:00

model_prx_pixel.py

fix hf telemetry and add prxpixel

2026-07-01 12:50:45 +02:00

model_prx.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_qwen.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_sana.py

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

model_sd3.py

enhance automated testing

2026-06-03 09:27:34 +02:00

model_sdxs.py

lint fix

2026-06-04 11:37:44 +02:00

model_segmoe.py

minor fixes from audit

2026-07-01 11:08:04 +02:00

model_stablecascade.py

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

model_step1x_edit.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

model_ultraflux.py

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

model_vibe.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

model_wanai.py

fix(video): apply Wan 2.2 MoE boundary at runtime in both image and video paths

2026-06-30 23:00:26 +01:00

model_xomni.py

refactor for consistent folders usage for all models

2026-06-08 08:24:21 +02:00

model_z_image.py

attempt to load incomplete model

2026-06-29 13:39:46 +02:00

model_zetachroma.py

automated pipeline registrations and tests

2026-06-04 10:18:09 +02:00

native_transformer.py

refactor(loader): collapse native-load logging to one summary line

2026-07-05 18:21:28 +01:00