github/InvokeAI
1
1
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI.git synced 2026-04-23 03:00:31 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix an edge case with model offloading from VRAM to RAM. If a GGML-quantized model is offloaded from VRAM inside of a torch.inference_mode() context manager, this will cause the following error: 'RuntimeError: Cannot set version_counter for inference tensor'.

Browse Source
This commit is contained in:
Ryan Dick
2025-01-07 15:59:50 +00:00
parent 67e948b50d
commit 85eb4f0312
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
invokeai/app/invocations/spandrel_image_to_image.py
View File

@@ -151,7 +151,7 @@ class SpandrelImageToImageInvocation(BaseInvocation, WithMetadata, WithBoard):
return pil_image
@torch.inference_mode()
@torch.no_grad()
def invoke(self, context: InvocationContext) -> ImageOutput:
# Images are converted to RGB, because most models don't support an alpha channel. In the future, we may want to
# revisit this.
@@ -197,7 +197,7 @@ class SpandrelImageToImageAutoscaleInvocation(SpandrelImageToImageInvocation):
description="If true, the output image will be resized to the nearest multiple of 8 in both dimensions.",
)
@torch.inference_mode()
@torch.no_grad()
def invoke(self, context: InvocationContext) -> ImageOutput:
# Images are converted to RGB, because most models don't support an alpha channel. In the future, we may want to
# revisit this.