The new engine that’ll power Easy Diffusion’s upcoming v4 release (i.e. sdkit3) has now been integrated into Easy Diffusion. It’s available to test by selecting v4 engine in the Settings tab (after enabling Beta). Please press Save and restart Easy Diffusion after selecting this.

It uses stable-diffusion.cpp and ggml under-the-hood, and produces optimized, lightweight builds for the target hardware.

The main benefits of Easy Diffusion’s new engine are:

1. Very lightweight - Less than 100 MB install footprint, compared to 3 GB+ for Forge and other PyTorch-based engines.

2. Much better for AMD/Intel/Integrated users - avoids the hot mess of ROCm and DirectML, by using a reliable Vulkan backend (that’s also used in llama.cpp).

3. Opportunity for even faster image generation in the future - this currently uses stock sd.cpp, which has room for further optimization.

4. Support for older GPUs - Vulkan supports older GPUs, especially older AMD GPUs unsupported by ROCm/PyTorch.

This supports:

• Windows, Linux, Mac

• NVIDIA, AMD, Intel GPUs (including integrated)

Users on Easy Diffusion’s Discord Server have already been trying it out, and it seems to work reasonably well. Atleast as well as I can expect for the first release of a fundamentally different engine for ED.

This is “future work”. It’s still missing a few features and has a few bugs (tracked here). It will take a while before it is ready for widespread use.

But for AMD users, ED4 is sometimes already better than ED 3.5/Forge (which uses ROCm/DirectML, and doesn’t work reliably for many AMD users). So that’s why ED4’s new engine has been fast-tracked for early release.