[Bug]: Stable diffusion model failed to load - cannot import name 'VQModelInterface' from 'ldm.models.autoencoder'

[ssalka]

Is there an existing issue for this?

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

What happened?

I started getting this error after upgrading a few extensions (control net, animate diff, dynamic prompts/wildcards, auto tls). I have been using the dev branch to get this fix and am on commit 5e80d9ee; I also tried on the the more recent commit 8aa51f68 and have done a full re-install by backing up my venv directory and letting the webui setup make a new one.

I checked in repositories\stable-diffusion-stability-ai\ldm\models\autoencoder.py, and did not see anything named VQModelInterface. Did something go wrong with my installation/setup?

Here is the relevant stack trace:

Loading VAE weights specified in settings: C:\stable-diffusion-webui\models\VAE\vae-ft-mse-840000-ema-pruned.ckpt
Applying attention optimization: xformers... done.
loading stable diffusion model: ImportError
Traceback (most recent call last):
  File "C:\Users\Me\AppData\Local\Programs\Python\Python310\lib\threading.py", line 973, in _bootstrap
    self._bootstrap_inner()
  File "C:\Users\Me\AppData\Local\Programs\Python\Python310\lib\threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "C:\Users\Me\AppData\Local\Programs\Python\Python310\lib\threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "C:\stable-diffusion-webui\modules\initialize.py", line 147, in load_model
    shared.sd_model  # noqa: B018
  File "C:\stable-diffusion-webui\modules\shared_items.py", line 112, in sd_model
    return modules.sd_models.model_data.get_sd_model()
  File "C:\stable-diffusion-webui\modules\sd_models.py", line 522, in get_sd_model
    load_model()
  File "C:\stable-diffusion-webui\modules\sd_models.py", line 655, in load_model
    sd_hijack.model_hijack.hijack(sd_model)
  File "C:\stable-diffusion-webui\modules\sd_hijack.py", line 259, in hijack
    import modules.models.diffusion.ddpm_edit
  File "C:\stable-diffusion-webui\modules\models\diffusion\ddpm_edit.py", line 27, in <module>
    from ldm.models.autoencoder import VQModelInterface, IdentityFirstStage, AutoencoderKL
ImportError: cannot import name 'VQModelInterface' from 'ldm.models.autoencoder' (C:\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\models\autoencoder.py)


Stable diffusion model failed to load

Steps to reproduce the problem

1. git checkout 5e80d9ee (dev branch)
2. Run webui-user.bat (using args --medvram --xformers)
3. See error preventing SD model load, hopefully

What should have happened?

The SD model should load & the UI setup should finish

Sysinfo

sysinfo-2023-11-24-17-19.txt

What browsers do you use to access the UI ?

Brave

Console logs

C:\stable-diffusion-webui>webui-user.bat
venv "C:\stable-diffusion-webui\venv\Scripts\Python.exe"
Python 3.10.6 (tags/v3.10.6:9c7b4bd, Aug  1 2022, 21:53:49) [MSC v.1932 64 bit (AMD64)]
Version: v1.6.0-298-g5e80d9ee
Commit hash: 5e80d9ee99c5899e5e2b130408ffb65a0585a62a
Launching Web UI with arguments: --medvram --xformers
[-] ADetailer initialized. version: 23.11.1, num models: 9
2023-11-24 09:13:42,199 - ControlNet - INFO - ControlNet v1.1.419
ControlNet preprocessor location: C:\stable-diffusion-webui\extensions\sd-webui-controlnet\annotator\downloads
2023-11-24 09:13:42,378 - ControlNet - INFO - ControlNet v1.1.419
Default key/cert pair was already generated by webui
Certificate trust store ready
Loading weights [0729a1570d] from C:\stable-diffusion-webui\models\Stable-diffusion\stable-diffusion-1_5.safetensors
Creating model from config: C:\stable-diffusion-webui\configs\v1-inference.yaml
Running on local URL:  https://127.0.0.1:7860

To create a public link, set `share=True` in `launch()`.
Startup time: 21.1s (prepare environment: 5.6s, import torch: 5.9s, import gradio: 1.6s, setup paths: 1.5s, initialize shared: 0.4s, other imports: 1.2s, setup codeformer: 0.2s, list SD models: 0.2s, load scripts: 3.3s, create ui: 0.8s, gradio launch: 0.5s).
Loading VAE weights specified in settings: C:\stable-diffusion-webui\models\VAE\vae-ft-mse-840000-ema-pruned.ckpt
Applying attention optimization: xformers... done.
loading stable diffusion model: ImportError
Traceback (most recent call last):
  File "C:\Users\Me\AppData\Local\Programs\Python\Python310\lib\threading.py", line 973, in _bootstrap
    self._bootstrap_inner()
  File "C:\Users\Me\AppData\Local\Programs\Python\Python310\lib\threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "C:\Users\Me\AppData\Local\Programs\Python\Python310\lib\threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "C:\stable-diffusion-webui\modules\initialize.py", line 147, in load_model
    shared.sd_model  # noqa: B018
  File "C:\stable-diffusion-webui\modules\shared_items.py", line 112, in sd_model
    return modules.sd_models.model_data.get_sd_model()
  File "C:\stable-diffusion-webui\modules\sd_models.py", line 522, in get_sd_model
    load_model()
  File "C:\stable-diffusion-webui\modules\sd_models.py", line 655, in load_model
    sd_hijack.model_hijack.hijack(sd_model)
  File "C:\stable-diffusion-webui\modules\sd_hijack.py", line 259, in hijack
    import modules.models.diffusion.ddpm_edit
  File "C:\stable-diffusion-webui\modules\models\diffusion\ddpm_edit.py", line 27, in <module>
    from ldm.models.autoencoder import VQModelInterface, IdentityFirstStage, AutoencoderKL
ImportError: cannot import name 'VQModelInterface' from 'ldm.models.autoencoder' (C:\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\models\autoencoder.py)


Stable diffusion model failed to load

Additional information

Currently enabled extensions:

[ssalka]

Solved my own problem, it was because the LDSR extension had somehow become disabled, I assume through some mechanism of the upgrade process though I don't understand why as I didn't target it for an upgrade and didn't turn it off manually. Found this out via this discussion