Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

No ONNX file found. Exporting ONNX… #54

Closed
bogus83 opened this issue Oct 18, 2023 · 9 comments
Closed

No ONNX file found. Exporting ONNX… #54

bogus83 opened this issue Oct 18, 2023 · 9 comments
Assignees

Comments

@bogus83
Copy link

bogus83 commented Oct 18, 2023

Can't "export default engine". I've tried 20 different checkpoints, all return the same error immediately:

Exporting realisticVisionV40_v40VAE to TensorRT
[INFO]: No ONNX file found. Exporting ONNX…
@FurkanGozukara
Copy link

it is accurate watch this

https://youtu.be/04XbtyKHmaE

@bogus83
Copy link
Author

bogus83 commented Oct 18, 2023

it is accurate watch this

https://youtu.be/04XbtyKHmaE

Sorry, I'm not paying to join your patreon.

@MorkTheOrk
Copy link
Collaborator

Does not look like a error, It notifies you that this checkpoint does not have a ONNX file yet. What are is the rest of the console print?

@MorkTheOrk MorkTheOrk self-assigned this Oct 18, 2023
@ValstrixMMM
Copy link

I just had this issue myself, but after some troubleshooting it seems that for whatever reason the commandline argument --medvram was causing the issue. After removing it and restarting, I was able to generate defaults. Not sure if you have that enabled or not, but it's worth a look.

@bogus83
Copy link
Author

bogus83 commented Oct 18, 2023

That's all that was displayed in the SD output. In the command prompt, it was throwing an error about "tensors on multiple devices". Looks like it was related to using --med-vram, because it did export correctly when I restarted SD without that flag.

@bogus83
Copy link
Author

bogus83 commented Oct 18, 2023

I just had this issue myself, but after some troubleshooting it seems that for whatever reason the commandline argument --medvram was causing the issue. After removing it and restarting, I was able to generate defaults. Not sure if you have that enabled or not, but it's worth a look.

Ha, you beat me to posting that by literally one second. :)
But yes, that does appear to have been the issue.

@MorkTheOrk
Copy link
Collaborator

Sorry for the confusion, we have added the --med-vram flag to the known issues into the readme. Are we good to close?

@bogus83
Copy link
Author

bogus83 commented Oct 18, 2023

I think so. Thanks for checking on this.

@maliang1234m
Copy link

我自己也遇到了这个问题,但是经过一些故障排除似乎无论出于何种原因,命令行参数 --medvram 都导致了该问题。删除它并重新启动后,我能够生成默认值。不确定您是否启用了该功能,但值得一看。

哈,你比我早一秒钟就发布了这个消息。:) 但是,是的,这似乎确实是问题所在。

Which file is it set in? Can you teach me something

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants