PNG Info no longer loads the correct model [Bug]:

[Ibycus66]

trafficstars

Is there an existing issue for this?

• [X] I have searched the existing issues and checked the recent builds/commits

What happened?

Not sure which version this started happening for me but when I use PNG info and send to txt2img or img2img the correct model no longer loads. It looks like automatic1111 has applied a new hash to my models and isn't identifying them properly.

Is anyone else having the same issue or know how I can fix it?

The setting/interface --->"When reading generation parameters " setting doesn't seem to matter

Steps to reproduce the problem

Load a PNG (from a few months ago) with info that uses a different model than what is currently loaded.

Send to txt2img

Note if the model changes automatically

Check model hash vs the PNG info hash for what that model should be

What should have happened?

PNG info use to load the correct model from older files when sending to txt2img

Commit where the problem happens

Commit hash: ea9bd9fc7409109adcd61b897abc2c8881161256

What platforms do you use to access the UI ?

Windows

What browsers do you use to access the UI ?

Google Chrome

Command Line Arguments

?__theme=dark

List of extensions

deforum-for-automatic1111-webui https://github.com/deforum-art/deforum-for-automatic1111-webui unknown depthmap2mask https://github.com/Extraltodeus/depthmap2mask.git unknown openOutpaint-webUI-extension https://github.com/zero01101/openOutpaint-webUI-extension unknown sd_dreambooth_extension https://github.com/d8ahazard/sd_dreambooth_extension unknown stable-diffusion-webui-depthmap-script https://github.com/thygate/stable-diffusion-webui-depthmap-script.git unknown stable-diffusion-webui-embedding-editor

Console logs

venv "H:\Stable_Diffusion_2\stable-diffusion-webui\venv\Scripts\Python.exe"
Python 3.10.7 (tags/v3.10.7:6cc6b13, Sep  5 2022, 14:08:36) [MSC v.1933 64 bit (AMD64)]
Commit hash: ea9bd9fc7409109adcd61b897abc2c8881161256
Installing requirements for Web UI



#######################################################################################################
Initializing Dreambooth
If submitting an issue on github, please provide the below text for debugging purposes:

Python revision: 3.10.7 (tags/v3.10.7:6cc6b13, Sep  5 2022, 14:08:36) [MSC v.1933 64 bit (AMD64)]
Dreambooth revision: 9f4d931a319056c537d24669cb950d146d1537b0
SD-WebUI revision: ea9bd9fc7409109adcd61b897abc2c8881161256

Checking Dreambooth requirements...
[+] bitsandbytes version 0.35.0 installed.
[+] diffusers version 0.10.2 installed.
[+] transformers version 4.25.1 installed.
[ ] xformers version N/A installed.
[+] torch version 1.12.1+cu113 installed.
[+] torchvision version 0.13.1+cu113 installed.

#######################################################################################################

Installing imageio-ffmpeg requirement for depthmap script
Installing pyqt5 requirement for depthmap script

Launching Web UI with arguments:
No module 'xformers'. Proceeding without it.
==============================================================================
You are running torch 1.12.1+cu113.
The program is tested to work with torch 1.13.1.
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.
==============================================================================
SD-Webui API layer loaded
Loading weights [39ffeb349b] from H:\Stable_Diffusion_2\stable-diffusion-webui\models\Stable-diffusion\V6-PieterV4-radaPoint5.ckpt
Creating model from config: H:\Stable_Diffusion_2\stable-diffusion-webui\configs\v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
Applying cross attention optimization (Doggettx).
Textual inversion embeddings loaded(0):
Model loaded in 4.9s (load weights from disk: 0.7s, create model: 0.4s, apply weights to model: 0.8s, apply half(): 0.9s, move model to device: 1.1s, load textual inversion embeddings: 0.8s).
Running on local URL:  http://127.0.0.1:7860

Additional information

No response

[nosdregamon]

I have the same issue on Windows with Firefox and Edge, my hash is ea9bd9f, too.

Thanks to Quicksettings (sd_model_checkpoint, sd_vae, CLIP_stop_at_last_layers, eta_noise_seed_delta, use_old_karras_scheduler_sigmas), I can see that ENSD "eta_noise_seed_delta" is also not filled it - it was before.

My plugins are the following: a1111-sd-webui-haku-img, a1111-sd-webui-tagcomplete, embedding-inspector, sd_dreambooth_extension, stable-diffusion-webui-instruct-pix2pix, stable-diffusion-webui-promptgen

I can also see: The Java-Script-Console throws an error, upon clicking the "send to txt2img"-button:

Uncaught (in promise) TypeError: L[jt[Gt]] is undefined ... ... B Button.svelte:9

Upon re-loading the page, I also get a lot of svelte-related messages (not errors, just message, but probably related) like

... [svelte-i18n] The message "🔄" was not found in "en-US", "en". ... [svelte-i18n] The message "Close" was not found in "en-US", "en". ...

[ewebgh33]

If this is going to be looked at, maybe we could set a checkbox in settings to do this, or not do this. Many times I want to load prompt settings for everything EXCEPT the model, because I want to try a different model. So the feature is good, but would be best if it could turned on/off in settings as each user likes. :)

[CCpt5]

I too have this issue, and hope the functionality returns to how it was in earlier versions. I put the setting to change (or not change) the model on the top bar using Quicksettings and that works fantasticly in older versions.

Thanks!

[VeMeth]

I have the same problem, rest of the info is carried over, only the model info is missing.

[AlUlkesh]

This PR should fix the issue: https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/7823