WebUI to convert ckpt/safetensor format model to ONNX format for AMD GPU or CPU for stable diffusion

preview_player
Показать описание
Thank you for watching! please consider to subscribe. thank you!
👉ⓢⓤⓑⓢⓒⓡⓘⓑⓔ

Affiliate links: buy on Amazon

In this video, the Webui for converting ckpt or safetensor format model to onnx format is shown.
For using the onnx format model in stable diffusion, the video likes are in
0:6 Intro
1:10 Scripts working directory, activate conda environment.
2:21 Run the webui and the interface layout
5:22 Example model converting
7:02 Memory RAM monitoring when converting in progress
8:50 Results
Рекомендации по теме
Комментарии
Автор

can you please add onx to ort converter web ui im trying to make some mobile models for stable diffusion on android no one has made a video on how to convert onnx model to ort

shareeftaylor
Автор

Does this work only with nvidia cards? I have amd card and when I click "submit" I get a lot of python errors.

thisperson
Автор

I was finally able to convert it, but why didn't you show to the end? I have now both folders "convert_diffuser_model" and "convert_model_onnx_format" but where do i have to put it? i tried to copy "convert_model_onnx_format" in the models/ONNX-Olive folder and renamed it to the name the safetensor file had before the conversion. It appears in the dropdown list, but gives an error when i try to create images. please show the whole process :(

PointlessClip
Автор

I get this error in automatic1111 when i tried using the dreamshaper model i converted using your script:



InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(float16)), expected: (tensor(float))

Kitten_Genocide
Автор

works. but can`t convert LORA files in safetensors format (from Civitai). may be you can add support? and add SDXL support;)

saymannskable
Автор

When I press submit I get a message all the time: something went wrong. please try another model file. with all files.

nolo
Автор

how hard would it be to add the ability to convert XL models?

shanold
Автор

ERROR: Could not install packages due to an OSError: [Errno 2] No such file or directory: wasn't able to install

mahmoudmajdoub
Автор

Maybe you have instructions how to use AMD GPU with UI version. I'm successfully using the console version, thanks to your previous video, but I don't know how to change the settings there. So I want to use ui.

san-chous
Автор

ModuleNotFoundError: No module named Bro can you help me to fix it?

kwk
Автор

I am getting black screen output with every model I try, even with stable diffusion. Why is that?

Nedrozak
Автор

Will this project ever support sdxl 0.9 & 1.0 ?

Myk
Автор

Brother can you make this for Google colab. I don't have a powerful machine

TReaterZZ
Автор

lot of issues with your converter. could you update it?

zerohcrows
Автор

Hey man! im getting

EnvironmentNameNotFound: Could not find conda environment: sd39
You can list all discoverable environments with `conda info --envs`.

when I run "conda activate sd39"

SuperOsba
Автор

Do you know of any ways of doing this for RVC?

foreverlostt
Автор

meh, can't run anything, such a shame with amd cards, so many errors that noone knows how to fix 😂

RestleSSOtaKU
Автор

i wish people could show how to do these with code rather than guis or scripts

JemiloII