Lightweight inference library for ONNX files, written in C++. It can run Stable Diffusion XL 1.0 on a RPI Zero 2 (or in 298MB of RAM) but also Mistral 7B on desktops and servers. ARM, x86, WASM, RISC-V supported. Accelerated by XNNPACK.
To see how the diffusion is progressing, it would be good to see diffused image stages in - for example - every single steps. The user could use paramter to show it or not. When showing like --showsteps without any extra parameter would show diffused image from every single steps. I know it would cause some extra cpu overhead, but would be really entertaining on more powerful systems than RPi.
I already started to thinking about how to achieve this, but other more talented developers might know this already.
To see how the diffusion is progressing, it would be good to see diffused image stages in - for example - every single steps. The user could use paramter to show it or not. When showing like --showsteps without any extra parameter would show diffused image from every single steps. I know it would cause some extra cpu overhead, but would be really entertaining on more powerful systems than RPi.
I already started to thinking about how to achieve this, but other more talented developers might know this already.
Update:
I found the way to achieve it.