Where is the code for the model checkpoint

[yqwjj]

I couldn't find this piece of code in ui.py of create_ui()

[w-e-w]

what do you mean?

---

ahh
if you're looking for the gradio element that creates the Stable Diffusion checkpoint dropdown you won't find it in the UI
this is because Stable Diffusion checkpoint is technically not a a part of of the main UI but a setting value, it only appears as if it's a UI because it is placed in quick settings by default

[yqwjj]

May I ask if you have any method? My idea is to obtain it through the components of gradio, but it cannot be implemented. I have also tried to obtain it through JavaScript, but I don't understand how to do it

[w-e-w]

I have understood the principle here through your prompt. However, I still cannot achieve my goal. I want to obtain the value of the user selection model dropdown box on the current page when clicking the submit button

there is no sd_model_checkpoint value when clicking on the submit button
as I have tried to explain it is technically a setting, a back-end setting, it is not a value that is submitted

as it is a setting, if you have multiple clients open you might experience a server client backend desync, so the reading the front end value is less meaningful
unless of course you do have a specific need of fetching frounend is displayed in the front end which in that case I guess you would need to use some sort of JavaScript

all settings can be obtained from the backend using python shared.opts.setting_key_name
so in this case it will be shared.opts.sd_model_checkpoint

but for checkpoint specifically it's more practical to look for information about it from shared.sd_model.sd_checkpoint_info

[yqwjj]

I understand what you mean. I want to create a feature that supports multiple users using it simultaneously. If I directly read the cache model, it cannot achieve the function of multiple users. That's why I obtained the dropdown menu selection value on the current page

[w-e-w]

currently the best method to make sure that you are using the model you wish is to utilize overwrite_settings

override_settings is used when loading png info you wull see a box

this will request a change of the model before image gen

you can also add sd_model_checkpoint a an extra option using the built-in extension extra-option-section

see wiki for more info https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/User-Interface-Customizations#additional-options-for-txt2img--img2img

this is also using override_setting internally

through API
example https://gist.github.com/w-e-w/0f37c04c18e14e4ee1482df5c4eb9f53#file-sd-webui-txt2img-img2img-api-example-py-L133

write your own script or extension that does this before processing
p.override_settings['sd_model_checkpoint'] = info.name

XYZ grid also uses this method to switch models

[yqwjj]

Thank you for your answer. I will try according to what you said