NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check.
Hey @JpKCS Thanks for the tip! You’re absolutely right, using batch count instead of batch size is definitely easier on the PC. I appreciate you pointing that out. Thanks for helping out, and keep those awesome suggestions coming!
When I attempt to use the python BAT file it comes up with this error. Installing torch and torchvision C:\Ai\stable-diffusion-webui\venv\Scripts\python.exe: No module named pip Traceback (most recent call last): File "C:\Ai\stable-diffusion-webui\launch.py", line 48, in main() File "C:\Ai\stable-diffusion-webui\launch.py", line 39, in main prepare_environment() File "C:\Ai\stable-diffusion-webui\modules\launch_utils.py", line 380, in prepare_environment run(f'"{python}" -m {torch_command}', "Installing torch and torchvision", "Couldn't install torch", live=True) File "C:\Ai\stable-diffusion-webui\modules\launch_utils.py", line 115, in run raise RuntimeError(" ".join(error_bits)) RuntimeError: Couldn't install torch. Command: "C:\Ai\stable-diffusion-webui\venv\Scripts\python.exe" -m pip install torch==2.1.2 torchvision==0.16.2 --extra-index-url download.pytorch.org/whl/cu121 Error code: 1 Press any key to continue . . .
ok so for anyone running into this I bring the solution, your bat file is trying to pull from a previous installation of python, so if you right click and edit the bat file, there's a path for the python version you installed. Put the location of your python 3.10.6 there, for example for me it looked like this: set PYTHON= C:\Users\user\AppData\Local\Programs\Python\Python310\python.exe Youre Welcome
Hi @musiccalgary, the developers of Automatic1111 officially recommend using Python 3.10.6 for the most stable experience. This version has been thoroughly tested and is most likely to work without issues. You might be able to run Automatic1111 with a newer version of Python, but it's generally not recommended as you could find comparability problems with key libraries on newer versions of Phyton. Thank you for watching!
![Stable Diffusion 1.5 - Windows Installation Guide [Tutorial]](http://i.ytimg.com/vi/ycQJDJ-qNI8/mqdefault.jpg)
Finally, an actual tutorial that covers everything LOL.
Thank you for watching!
NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check.
Thanks, it worked!
Glad it helped, Thank you for watching!
a very good tutorial
Hi dear, thanks for the tutorial, but I did open the webui-user and wait until it finished It didn't give any web address.
You shouldn't set batch size to create multiple variations, you should use batch count instead - it's easier for your PC! 🚨
Hey @JpKCS Thanks for the tip! You’re absolutely right, using batch count instead of batch size is definitely easier on the PC. I appreciate you pointing that out. Thanks for helping out, and keep those awesome suggestions coming!
When I attempt to use the python BAT file it comes up with this error.
Installing torch and torchvision
C:\Ai\stable-diffusion-webui\venv\Scripts\python.exe: No module named pip
Traceback (most recent call last):
File "C:\Ai\stable-diffusion-webui\launch.py", line 48, in
main()
File "C:\Ai\stable-diffusion-webui\launch.py", line 39, in main
prepare_environment()
File "C:\Ai\stable-diffusion-webui\modules\launch_utils.py", line 380, in prepare_environment
run(f'"{python}" -m {torch_command}', "Installing torch and torchvision", "Couldn't install torch", live=True)
File "C:\Ai\stable-diffusion-webui\modules\launch_utils.py", line 115, in run
raise RuntimeError("
".join(error_bits))
RuntimeError: Couldn't install torch.
Command: "C:\Ai\stable-diffusion-webui\venv\Scripts\python.exe" -m pip install torch==2.1.2 torchvision==0.16.2 --extra-index-url download.pytorch.org/whl/cu121
Error code: 1
Press any key to continue . . .
me too
what version of python do u have?
ok so for anyone running into this I bring the solution, your bat file is trying to pull from a previous installation of python, so if you right click and edit the bat file, there's a path for the python version you installed. Put the location of your python 3.10.6 there, for example for me it looked like this: set PYTHON= C:\Users\user\AppData\Local\Programs\Python\Python310\python.exe
Youre Welcome
I had this problem, but then I realized that I installed version 3.13. The video clearly says to install 3.10.6. Fixed.
Hello. You can do this installation on a laptop with Intel(R) Core(TM) i7-3537U CPU @ 2.00GHz 2.50 GHz and 8 RAM. Thank you
to be fair - the most important is GPU if you don't have one it will most likely be unusable.
hey how can I runn it on my intel gpu ?
What if you don't want to install old python?
Hi @musiccalgary, the developers of Automatic1111 officially recommend using Python 3.10.6 for the most stable experience. This version has been thoroughly tested and is most likely to work without issues. You might be able to run Automatic1111 with a newer version of Python, but it's generally not recommended as you could find comparability problems with key libraries on newer versions of Phyton. Thank you for watching!
Hello. You can do this installation on a laptop with Intel(R) Core(TM) i7-3537U CPU @ 2.00GHz 2.50 GHz and 8 RAM. Thank you