🤯 2X SPEED BOOST FOR STABLE DIFFUSION | HUGE NEW NEWS |

Watch the video:

0:00 - 2x Speed improvement in Stable Diffusion announcement
1:17 - Downloading latest Nvidia driver
1:47 - About new ONNX/Olive format
2:50 - What about using ONNX/Olive in Automatic1111/WebUI?
4:50 - Set up ONNX/Olive
6:07 - Optimizing/Converting models to ONNX/Olive
6:40 - Testing ONNX/Olive Speed (INSANE)
7:20 - ONNX vs normal Stable Diffusion (2x the speed!)
7:50 - Optimized Automatic1111 launch commands vs ONNX (1.5x Speed!)
8:38 - Trying to get ONNX to work in SDUI
10:15 - Stable Diffusion XUI (With ONNX support!)
13:17 - Summary

Nvidia has announced HUGE news: 2x improvement in speed for Stable Diffusion and more with the latest driver. Using it is a little more complicated, but the speed boost is there! Exciting things coming in the future of AI. This video covers installing and using the new ONNX/Olive models and converter, as well as converting models, generating images using the new format, as well as places where you can use it now - and where you can’t. Support will need to grow and be added to programs to use this new format.

Blog announcement: https://blogs.nvidia.com/blog/2023/05/23/microsoft-build-nvidia-ai-windows-rtx/
DirectML devblog: https://devblogs.microsoft.com/directx/dml-stable-diffusion/
Nvidia driver download: https://www.nvidia.com/download/index.aspx

Olive GitHub (Install commands): https://github.com/microsoft/Olive/tree/main/examples/directml/stable_diffusion

Stable Diffusion XUI: https://github.com/ForserX/StableDiffusionUI

stable-diffusion-webui-directml (Coudn’t get tow work): https://github.com/lshqqytiger/stable-diffusion-webui-directml