Alexander Eichhorn 4aa87f57a1 fix(ernie-image): stop warning about a missing scheduler for GGUF files ...

Only a Diffusers folder is expected to carry scheduler/, so single files of any format no longer warn.
The GGUF forward test compared all-zero outputs (diffusers zero-inits adaLN and the final layer); it now fills
those in and also covers Q4_K, the starters' type, on a 256-wide geometry.

2026-10-10 23:36:15 +02:00

..

anima

fix(anima): cast the whole semantic connector under FP8 Storage

2026-10-02 04:43:45 +02:00

architectures

feat(ui): reliability pass plus canvas, gallery and frontend layout features (#9682)

2026-10-08 09:08:17 -04:00

ernie_image

feat: ernie image/turbo (#9115)

2026-07-31 00:51:43 +00:00

flux

Merge branch 'main' into fix/flux-schedule-shift-clamp

2026-09-29 20:31:47 +02:00

flux2

fix(flux2): img2img posterization - noise scaling and sigma schedule (#9556)

2026-09-23 21:17:18 -04:00

ideogram4

fix(tests): stabilize browser interactions and Windows dtype checks

2026-09-29 16:43:08 -06:00

image_util

feat(db): move every query into one database layer and support MySQL and MariaDB (#9702)

2026-10-09 22:40:06 -06:00

ip_adapter

refactor: model manager v3 (#8607)

2025-10-15 10:18:53 +11:00

krea2

chore: merge upstream/main

2026-09-24 16:13:11 -06:00

llava_onevision

feat(prompt): show phase + token progress for LLM prompt expansion (#9204)

2026-07-31 04:08:33 +00:00

ltx2

Merge branch 'main' into feat/ltx2-a2v-first-frame

2026-10-07 21:35:02 -04:00

minimax_h3

fix(deps): resolve high and critical dependabot alerts (#9723)

2026-10-09 23:50:30 +00:00

model_manager

fix(ernie-image): stop warning about a missing scheduler for GGUF files

2026-10-10 23:36:15 +02:00

patches

feat(anima): apply Anima LoRAs and LLLite adapters to the blocks the finetunes kept

2026-10-02 04:31:26 +02:00

pid

fix(pid): make PiD 1.5 usable for FLUX.1 and FLUX.2

2026-10-10 01:09:22 +02:00

quantization

Merge branch 'main' into feat/rocm10

2026-10-10 12:30:03 -04:00

qwen2_5_vl

feat(models): bundle encoder tokenizers and configs instead of fetching them

2026-09-23 20:15:34 +02:00

qwen3

feat(models): bundle encoder tokenizers and configs instead of fetching them

2026-09-23 20:15:34 +02:00

qwen3_5

feat(anima): support Anima-2.9B and Anima-3.8B finetunes

2026-10-02 00:53:56 +02:00

qwen3_vl

feat(models): bundle encoder tokenizers and configs instead of fetching them

2026-09-23 20:15:34 +02:00

rectified_flow

feat(model): Add ER SDE / DPM++ 2M Scheduler Support For Anima (#9125)

2026-05-08 02:59:23 +00:00

stable_diffusion

Fix arguments for hidiffusion hook (#9679)

2026-10-04 22:21:23 +00:00

t5

feat: t5 encoder gguf support (#9324)

2026-07-24 01:34:47 +00:00

text_llm

feat(gpu): run prompt tools and image embedding on an idle GPU

2026-09-27 12:04:20 -04:00

tiles

In CalculateImageTilesEvenSplitInvocation to have overlap_fraction becomes just overlap. This is now in pixels rather than as a fraction of the tile size.

2023-12-17 15:10:50 +00:00

util

fix(rocm): hide the Windows iGPU whatever the device settings say, and bound the probe

2026-10-09 05:19:14 +02:00

wan

test(rocm): start the conv3d patch idempotence tests from an unpatched class

2026-09-19 18:31:33 -04:00

z_image

feat(models): bundle encoder tokenizers and configs instead of fetching them

2026-09-23 20:15:34 +02:00

__init__.py

POC of a test that depends on models.

2023-10-05 15:35:58 -04:00

test_clip_text_model_wrapper.py

fix(deps): resolve high and critical dependabot alerts (#9723)

2026-10-09 23:50:30 +00:00

test_text_llm_pipeline.py

feat(prompt): show phase + token progress for LLM prompt expansion (#9204)

2026-07-31 04:08:33 +00:00