-
Begin to load 1 model
[Unload] Trying to free 4495.77 MB for cuda:0 with 0 models keep loaded ...
[Unload] Current free memory is 2297.73 MB ...
[Unload] Unload model KModel
[Memory Management] Cu…
-
C:\forge\webui\extensions-builtin\sd_forge_kohya_hrfix\scripts\kohya_hrfix.py
Traceback (most recent call last):
File "C:\forge\webui\modules\scripts.py", line 892, in process_before_ever…
-
**Unable to convert from tflite to kmodel**
I am trying to export a custom model to kmodel format using nncase for k210. I am able to convert my model to tflite successfully but conversion to kmodel …
-
I'm doing some demanding X/Y/Z plots to ascertain the best combinations of checkpoint, sampler, and scheduler to use with a new Lora I'm creating. Most commonly, at seemingly random points, image gen…
-
each time this take around 10-15 seconds while image generation itself take 30 sec for 20 step on a 3080 ti card fp8 quant flux
this happens each time if i generate in batch or single single
M…
-
**Describe the bug**
Unhandled exception. System.InvalidOperationException: Only The Expr Have CheckedType Can Get It's Shape
at Nncase.IR.Expr.get_CheckedShape()
at Nncase.Util.ShapeIndex(Ex…
-
我用一个sd1.5模型生成9张图片,内存占用基本维持在这个值:
I used an SD1.5 model to generate 9 pictures, and the memory usage basically remained at this value:
I say right away, all other models are working fine.
Up to 97% of the generation, the image looks q…
-
Hi I had to roll back to commit 720b80da due to flux.1 dev consuming more memory than the previous version. With the current version it seems to want to load ~32Gigs and winds up swapping with a simpl…
-
### Checklist
- [X] The issue exists after disabling all extensions
- [X] The issue exists on a clean installation of webui
- [ ] The issue is caused by an extension, but I believe it is caused by a …