Open KintCark opened 1 month ago
I looked it over It looks pretty straight forward.
have you considered learning to code?
I work weekends but if you check back in a week, I might be able to run it.
I'm wondering why the local diffusion still uses ONNX when it's so slow and doesn't appear to support lora's and img2img. The app is great
but implementing qualcomm's diffusion library or google's on device mediapipe would potentially change the entire system.
I looked it over It looks pretty straight forward.
have you considered learning to code?
I work weekends but if you check back in a week, I might be able to run it.
I'm wondering why the local diffusion still uses ONNX when it's so slow and doesn't appear to support lora's and img2img. The app is great
but implementing qualcomm's diffusion library or google's on device mediapipe would potentially change the entire system.
If u figure out how to use the Google colab notebook can u please show me in a video?
Is there a way to convert safetensors to ort format? Do you know how to use the Google colab notebook? The guy that made all those models to ort format used it but he wouldn't tell me the exact details on how to use it step by step, we can make more models but I'm not that smart when it comes to Google colab:https://github.com/Fcucgvhhhvjv/Android-Stable-diffusion-ONNX/blob/master/Full_(fixed)(gpu)(custom)(withsafetensor)torch_2_1_premodel_conversion_script.ipynb