Dave-Swagten
commented
10 months ago I have the same problem on a new installation running win11 using 7900xtx
Open Infamousfish opened 10 months ago
Dave-Swagten
commented
10 months ago I have the same problem on a new installation running win11 using 7900xtx
k0zer0g
commented
10 months ago same problem after installing the new version win11 rx580
dsthayl
commented
10 months ago Same error. Fresh install on a RX590, worked perfectly some months ago, 512x512 tooked only 15 or so seconds. Now cant even load settings or any model at all.
Infamousfish
commented
10 months ago I found a "solution" https://youtu.be/O40W-VOx5q0
roxas1212
commented
10 months ago I found a "solution" https://youtu.be/O40W-VOx5q0
how to use this for google colab?
EKKIV0K
commented
10 months ago did you tried to update python ? for the 3.10.9 version instead of 3.10.6 ? 2nd, did you tried to delete the virtual environement on the root of your a1111 folder ? please try this delete the "venv" folder and then click on webui to start a1111 as usual.
this will reinstall the VENV folders (no worries).
Infamousfish
commented
10 months ago did you tried to update python ? for the 3.10.9 version instead of 3.10.6 ? 2nd, did you tried to delete the virtual environement on the root of your a1111 folder ? please try this delete the "venv" folder and then click on webui to start a1111 as usual.
this will reinstall the VENV folders (no worries).
I appreciate your comment but I got to work eventually without doing any of what do you said but I couldn't use hires fix so I'm sticking to 1.6.1 for now until there's a new update
roxas1212
commented
10 months ago did you tried to update python ? for the 3.10.9 version instead of 3.10.6 ? 2nd, did you tried to delete the virtual environement on the root of your a1111 folder ? please try this delete the "venv" folder and then click on webui to start a1111 as usual. this will reinstall the VENV folders (no worries).
I appreciate your comment but I got to work eventually without doing any of what do you said but I couldn't use hires fix so I'm sticking to 1.6.1 for now until there's a new update
How do I downgrade the version to 1.6.1
Infamousfish
commented
10 months ago did you tried to update python ? for the 3.10.9 version instead of 3.10.6 ? 2nd, did you tried to delete the virtual environement on the root of your a1111 folder ? please try this delete the "venv" folder and then click on webui to start a1111 as usual. this will reinstall the VENV folders (no worries).
I appreciate your comment but I got to work eventually without doing any of what do you said but I couldn't use hires fix so I'm sticking to 1.6.1 for now until there's a new update
How do I downgrade the version to 1.6.1
I linked a video above
roxas1212
commented
10 months ago did you tried to update python ? for the 3.10.9 version instead of 3.10.6 ? 2nd, did you tried to delete the virtual environement on the root of your a1111 folder ? please try this delete the "venv" folder and then click on webui to start a1111 as usual. this will reinstall the VENV folders (no worries).
I appreciate your comment but I got to work eventually without doing any of what do you said but I couldn't use hires fix so I'm sticking to 1.6.1 for now until there's a new update
How do I downgrade the version to 1.6.1
I linked a video above
When I run "git checkout 03eec1791be011e087985ae93c1f66315d5a250e"
It appears "fatal: reference is not a tree: 03eec1791be011e087985ae93c1f66315d5a250e"
EKKIV0K
commented
10 months ago To downgrade stable diffusion just download the version 1.6.1 on the dedicated repo :)
Don't git clone ! Juste download the zip and install it like usual :)
roxas1212
commented
10 months ago Do you mean to modify the code in it? Can you help me modify the ipynb file above? I'm not familiar with COLAB and reprogramming. I'm really clueless, thanks
EKKIV0K
commented
10 months ago Nono here is the link to the correct repo :
No link available for 1.6.1 didn't found it. (No time il on the hurry right now)
Juste DOWNLOAD it and not GIT CLONE :) cause if you git clone it it will install the last version (1.7.0)
I tried the 1.7.0 and no problem using hiresfix.
Envoyé à partir de Outlook pour Androidhttps://aka.ms/AAb9ysg
From: roxas1212 @.> Sent: Friday, February 9, 2024 3:19:06 PM To: AUTOMATIC1111/stable-diffusion-webui @.> Cc: EKKIVOK @.>; Comment @.> Subject: Re: [AUTOMATIC1111/stable-diffusion-webui] [Bug]:Issue after install new version of webui (Issue #14859)
Do you mean to modify the code in it? Can you help me modify the ipynb file above? I'm not familiar with COLAB and reprogramming. I'm really clueless, thanks
— Reply to this email directly, view it on GitHubhttps://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/14859#issuecomment-1936013140, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A4TH3VYZBU2QJKDOHMBTFDDYSYV5VAVCNFSM6AAAAABC5YBOR6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMZWGAYTGMJUGA. You are receiving this because you commented.Message ID: @.***>
Infamousfish
commented
9 months ago Nono here is the link to the correct repo : https://github.com/AUTOMATIC1111/stable-diffusion-webui/tree/5ef669de080814067961f28357256e8fe27544f4 No link available for 1.6.1 didn't found it. (No time il on the hurry right now) Juste DOWNLOAD it and not GIT CLONE :) cause if you git clone it it will install the last version (1.7.0) I tried the 1.7.0 and no problem using hiresfix. Envoyé à partir de Outlook pour Androidhttps://aka.ms/AAb9ysg … ____ From: roxas1212 @.> Sent: Friday, February 9, 2024 3:19:06 PM To: AUTOMATIC1111/stable-diffusion-webui @.> Cc: EKKIVOK @.>; Comment @.> Subject: Re: [AUTOMATIC1111/stable-diffusion-webui] [Bug]:Issue after install new version of webui (Issue #14859) Do you mean to modify the code in it? Can you help me modify the ipynb file above? I'm not familiar with COLAB and reprogramming. I'm really clueless, thanks — Reply to this email directly, view it on GitHub<#14859 (comment)>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A4TH3VYZBU2QJKDOHMBTFDDYSYV5VAVCNFSM6AAAAABC5YBOR6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMZWGAYTGMJUGA. You are receiving this because you commented.Message ID: @.***>
Do you use amd or nvidia?
roxas1212
commented
9 months ago Thank you for helping me, who is ignorant, I have successfully downgraded to 1.6.1 Although I haven't solved the problem that occurred in the older version. But I've been able to use it temporarily. https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/14854 If anyone can help me solve it please let me know. Because I always like to work on this version, thanks.
EKKIV0K
commented
9 months ago Thank you for helping me, who is ignorant, I have successfully downgraded to 1.6.1 Although I haven't solved the problem that occurred in the older version. But I've been able to use it temporarily. #14854 If anyone can help me solve it please let me know. Because I always like to work on this version, thanks.
can you post a log ? or a screenshot when your python script loaded a1111 ? it will help to get more informations about the probleme occured.
roxas1212
commented
9 months ago It can generate links normally
Python 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]
Commit hash: e5be9d1a5cee6e189ae7b86e9b968c1bf74ea960
Installing requirements
Launching Web UI with arguments: --theme dark --share --gradio-debug --disable-safe-unpickle --no-half-vae --reinstall-xformers --enable-insecure-extension-access --opt- channelslast
2024-02-10 09:30:47.844581: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2024-02-10 09:30:47.844635: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2024-02-10 09:30:47.846096: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2024-02-10 09:30:47.854012: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-02-10 09:30:49.128411: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
No module 'xformers'. Proceeding without it.
================================================== ============================
You are running torch 1.13.1+cu117.
The program is tested to work with torch 2.0.0.
To reinstall the desired version, run with commandline flag --reinstall-torch.
Beware that this will cause a lot of large files to be downloaded, as well as
there are reports of issues with training tab on the latest version.
Use --skip-version-check commandline argument to disable this check.
================================================== ============================
Checkpoint darkSushiMixMix_darkerPruned.safetensors [fb44463063] not found; loading fallback v1-5-pruned-emaonly.safetensors [6ce0161689]
Loading weights [6ce0161689] from /content/test1/models/Stable-diffusion/v1-5-pruned-emaonly.safetensors
Creating model from config: /content/test1/configs/v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
Couldn't find VAE named vae-ft-mse-840000-ema-pruned.ckpt; using None instead
Applying cross attention optimization (Doggettx).
Textual inversion embeddings loaded(2): badhandv4, EasyNegative
Model loaded in 3.8s (create model: 0.6s, apply weights to model: 0.7s, apply channels_last: 0.7s, apply half(): 0.4s, load VAE: 0.6s, move model to device: 0.6s).
Running on local URL: http://127.0.0.1:7860
Running on public URL: https://b8399fba8ab3ba2f61.gradio.liveBut when I enter the link, it says ERROR and doesn't work.
EKKIV0K
commented
9 months ago @roxas1212
Seems you run an extra outdated version of SD. since, extensions was updated a lot can create issues with this version. you are using python 3.10.12 wich is better to use 3.10.9. it seems to run on your cpu (extra slow) but you have Cuda installed (then it seems your systems can handle cuda, you have nvidia card right ?) Cause i see that your tensorflow is optimised for cpu wich he NEED to be optimised for GPU. (wich can handle torch 2.0, by the way..)
i see you run an old version of torch (1.13.1 and cuda 11.7) there was a lot of update since, toche must be updated. try the cmd args "--reinstall-torch".
for your models it seems you installed all of them on another hard drive right ? But in the configuration tab of your SD the adress of them are not correct.
try this arguments as a command on your Webui.bat :
--ckpt-dir "F:\SDXL Models" --lora-dir "J:\Fooocus_win64_2-1-25\Fooocus\models\lora XL"
(this my personal links, replace them by your own)
your outputs directory are wrong too onnx cannot have multiple HDD like : c:/, e:/ etc..... just use one Disk drive like C. For the other one, just use the default, or simple create another links on a the "image folder" (on windows) and then create the folder for each things. and then put the correct link into the UI.
And, Voila :)
roxas1212
commented
9 months ago @roxas1212
Seems you run an extra outdated version of SD. since, extensions was updated a lot can create issues with this version. you are using python 3.10.12 wich is better to use 3.10.9. it seems to run on your cpu (extra slow) but you have Cuda installed (then it seems your systems can handle cuda, you have nvidia card right ?) Cause i see that your tensorflow is optimised for cpu wich he NEED to be optimised for GPU. (wich can handle torch 2.0, by the way..)
i see you run an old version of torch (1.13.1 and cuda 11.7) there was a lot of update since, toche must be updated. try the cmd args "--reinstall-torch".
for your models it seems you installed all of them on another hard drive right ? But in the configuration tab of your SD the adress of them are not correct.
try this arguments as a command on your Webui.bat :
--ckpt-dir "F:\SDXL Models" --lora-dir "J:\Fooocus_win64_2-1-25\Fooocus\models\lora XL"
(this my personal links, replace them by your own)
your outputs directory are wrong too onnx cannot have multiple HDD like : c:/, e:/ etc..... just use one Disk drive like C. For the other one, just use the default, or simple create another links on a the "image folder" (on windows) and then create the folder for each things. and then put the correct link into the UI.
And, Voila :)
I can't believe my old version is working again. Thank you so much.
mikedebian
commented
6 months ago I have the same issue on Arch Linux. The WebUI loads but nothing works and the image looks like the one in the original post. It errors out with the same error messages on the latest release. Even the settings page is weird and don't load properly. This is with a fresh install. Can't select any checkboxes and it all looks "off" (No I'm not suffering from the Chinese CSS bug).
After having checked out the commit that Infamousfish posted, it worked again, however this is quite an old version. I can't find any reference about this bug being worked on.
Checklist
What happened?
after installing webui I get an error which doesn't let me use it, I had the old webui which was working fine but I decide to do a fresh install and it's not working anymore
Steps to reproduce the problem
What should have happened?
For it to launch and work like before
What browsers do you use to access the UI ?
Mozilla Firefox
Sysinfo
Internal Server Error
Console logs
Additional information
win11 6700xt