You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I tried using the SDXL. 2 models.
1: hassakuXLIllustrious_v22.safetensors
won't load.
Loading model......
Maybe you are using: SDXL
Unhandled exception. System.InvalidOperationException: Sequence contains no matching element
at System.Linq.ThrowHelper.ThrowNoMatchException()
at System.Linq.Enumerable.First[TSource](IEnumerable`1 source, Func`2 predicate)
at StableDiffusionSharp.ModelLoader.SafetensorsLoader.LoadSafetensors(Module module, String fileName, String maybeAddHeaderInBlock) in D:\temp\StableDiffusionSharp\StableDiffusionSharp\ModelLoader\SafetensorsLoader.cs:line 133
at StableDiffusionSharp.ModelLoader.ModelLoader.LoadModel(Module module, String fileName, String maybeAddHeaderInBlock) in D:\temp\StableDiffusionSharp\StableDiffusionSharp\ModelLoader\ModelLoader.cs:line 19
at StableDiffusionSharp.Modules.SDModel.LoadModel(String modelPath, String vaeModelPath, String vocabPath, String mergesPath) in D:\temp\StableDiffusionSharp\StableDiffusionSharp\Modules\SDModel.cs:line 107
at StableDiffusionSharp.StableDiffusion.LoadModel(String modelPath, String vaeModelPath, String vocabPath, String mergesPath) in D:\temp\StableDiffusionSharp\StableDiffusionSharp\StableDiffusion.cs:line 50
at StableDiffusionDemo_Console.Program.Main(String[] args) in D:\temp\StableDiffusionSharp\StableDiffusionDemo_Console\Program.cs:line 40
line 40:
sd.LoadModel(sdModelPath);
sdModelPath is correct!
dreamshaperXL_alpha2Xl10.safetensors
It seems to work, but I didn't wait the result. The frame generation speed is about 20 times slower than in Python in standard webui or, for example, sd.reForge.
In hassakuXLIllustrious_v22.safetensors, it seems that “conditioner.embedders.0.transformer.text_model.embeddings.position_ids” block is missing. The block sometimes is like a tensor with shape of [1,77], float16 and value from 0 to 76, but sometimes can be other values.
In SDXL, it will take about 9G VRAM for 1024x1024 image, maybe your VRAM is less than that, so it will takes a lot of time. Because in SD-webui or other GUIs have low-VRAM mode, so it will take less VRAM. Low-VRAM mode will be added later.
Activity
IntptrMax commentedon May 12, 2025