mirror of
https://github.com/vladmandic/automatic
synced 2026-08-31 01:20:59 +02:00
CalamitousFelicitousness
3368ce95f7
Qwen-Image: diffusers registers a no-op identity lambda for QwenImageTransformer2DModel; from_single_file silently accepted any key naming and loaded with mismatches. QWEN_SPEC sets converter=None to short-circuit the no-op. Custom transformer classes (Lens, Step1X-Edit, VIBE Sana, f-lite): runtime-registered onto diffusers, never in SINGLE_FILE_LOADABLE_CLASSES. Specs land in each arch's existing pipelines/<arch>/__init__.py. NextStep skipped: loader returns None up front.
16 lines
473 B
Python
16 lines
473 B
Python
from pipelines.native_transformer import TransformerSpec
|
|
from pipelines.step1x.pipeline_output import Step1XEditPipelineOutput
|
|
from pipelines.step1x.pipeline_step1x_edit import Step1XEditPipeline
|
|
from pipelines.step1x.transformer_step1x_edit import Step1XEditTransformer2DModel
|
|
|
|
|
|
STEP1X_SPEC = TransformerSpec(cls=Step1XEditTransformer2DModel)
|
|
|
|
|
|
__all__ = [
|
|
"STEP1X_SPEC",
|
|
"Step1XEditPipeline",
|
|
"Step1XEditPipelineOutput",
|
|
"Step1XEditTransformer2DModel",
|
|
]
|