HonZuna

Results 6 comments of HonZuna

I cant use n-gpu-layers since snapshot-2024-04-28

> Yeah, upgrade cuda to 12.1. No change at all, still the same with 12.1.1 (i checked PATH its correctly to new folder) ``` 16:20:52-424359 INFO Loading "Fimbulvetr-11B-v2-Test-14.q8_0.gguf" 16:20:52-456863 INFO...

I cant use n-gpu-layers since snapshot-2024-04-28

I just did, good point. But unfortunately it does not help either. ``` 01:50:06-213795 INFO Loading "Fimbulvetr-11B-v2-Test-14.q8_0.gguf" 01:50:06-267801 INFO llama.cpp weights detected: "models\Fimbulvetr-11B-v2-Test-14.q8_0.gguf" llama_model_loader: loaded meta data with 22 key-value...

I cant use n-gpu-layers since snapshot-2024-04-28

> I had the same problem in the last days (on a Linux system). Now I executed `update_wizard_linux.sh` and afterwards the GPU was used again. Try to execute `update_wizard_windows.bat` Thank...

[Feature Request]: LoRA-weight slider

Good idea.

Sometimes if i click the preview image before its finished the preview sticks and finished renders will no longer be shown requiring reboot/refresh.

Same issue for me. I have no idea how refresh does not help but new window/tab does.

[FEATURE_REQUEST] Story Writing/Text Editor Mode

Great idea.