Qing

Hi, which model are you using? This error might be due to an error during the model switching process. Could you try using this model directly at startup?

Hi, I currently don’t feel very motivated to update this project (because I’ve seen many new Gen AI projects with much better results, like the cleanup feature built into Samsung...

please check this: https://www.iopaint.com/file_manager When starting the service, you can add a `--mask-dir` option, and then on the frontend, there can be a page to select mask from the folder.

抱歉最近没有太多时间维护，有时间了我复现试试。

Multiple models are not supported. `--model=Sanster/PowerPaint-V1-stable-diffusion-inpainting` is placed after `--model=lama`, so `PowerPaint-V1-stable-diffusion-inpainting` is used when the service starts. You can switch models in the settings on the web page.

Hi, thank you for supporting this project. I watched the video in your issue, and it seems the problem might be caused by a proxy, network issues, or possibly a...

[installer_config.json](https://github.com/user-attachments/files/20299587/installer_config.json) `win_config` will save a JSON configuration file. You can use this configuration file (placed in the same directory as win_config) and try starting the service. If it works, you...

According to the logs, the service started normally. Please refer to the content of this issue regarding the white screen problem. https://github.com/Sanster/IOPaint/issues/460#issuecomment-1960656824

Please make sure `big-lama.pt` file exists, You can manually download it here https://github.com/Sanster/models/releases/download/add_big_lama/big-lama.pt

An explantion of buffer(ImagePool) purpose: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/issues/75