ssbugman

Instruction very confusing with this text which is not a cmd text and not a valid dockerfile. Code itself broken and contain lot of errors. "attempted relative import beyond top-level...

Incompatible with new GPU

4

GeForce RTX 3080 with CUDA capability sm_86 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 compute_37 New...

Error about mistake in code

2

Traceback (most recent call last): File "P:\SimSwapmain\test_one_image.py", line 30, in opt = TestOptions().parse() File "P:\SimSwapmain\options\base_options.py", line 84, in parse torch.cuda.set_device(self.opt.gpu_ids[0]) File "C:\Python310\lib\site-packages\torch\cuda\__init__.py", line 313, in set_device torch._C._cuda_setDevice(device) AttributeError: module 'torch._C'...

all LORAs produce "modules.devices.NansException: A tensor with all NaNs was produced in Unet" in StableDiffusion

1

Was using very old Kohya version, updated with fresh install to be able to train SDXL and all new LORA become broken. Backup copy of old version also stopped working...

Nothing happens following instruction.

15

I did all manual steps and getting just blender cube render on F12. Activating script just print new "bpy.ops.text.run_script()" line in tab under Blender terminal window. Seems manual missing important...

Аномальное пожирание памяти браузером

1

### Operating system / операционная система Windows 10 ### Running as service / Запуск программы как сервис I run it as a regular program / Запускаю программу обычным образом ###...

Lora not working after update

1

activating extra network lora with arguments [, , , ]: AttributeError Traceback (most recent call last): File "V:\webui_forge_cu121_torch231\webui\modules\extra_networks.py", line 135, in activate extra_network.activate(p, extra_network_args) File "V:\webui_forge_cu121_torch231\webui\extensions-builtin\Lora\extra_networks_lora.py", line 43, in activate...

Lora models list broken after update

2

When pushing refresh button lora models not show up, instead error appears: *** Error loading embedding lycoris_challenge_v1.safetensors Traceback (most recent call last): File "V:\webui_forge_cu121_torch231\webui\modules\textual_inversion\textual_inversion.py", line 200, in load_from_dir self.load_from_file(fullfn, fn)...

RAM leak on starting generation - model double loaded into RAM again on each generation after first

1

Clean install. First generation works fine. When pushing Generate second time - model loaded into RAM again which instantly cause RAM overflow with Flux models. After generation Forge not unloading...

After flux update merging models stoped working. "Error merging checkpoints: 'NoneType' object has no attribute 'keys'" Now it stll not fixed, now it become even worse. On pushing merge button...