In this video I’ll be demoing Stable Diffusion running locally on an RX 6900 XT. The key to good iterations per second is high VRAM and lots of compute. ROCm provides an alternative to CUDA for ML workloads. This applies to training LLMs and various AI application use.
Join the Discord if you are interested in PC DIY and all things tech
Full A.I. Workstation Specs:
CPU: AMD Ryzen 7 7700X -
Cooler: Noctua NH-D15 -
Motherboard: Gigabyte X670 AORUS Elite AX -
RAM: Corsair Vengeance DDR5 64GB 5200mhz -
GPU: AMD Radeon RX 6900 XT -
PSU: MEG Ai1300P PCIE5 PSU -
Case: Fractal Torrent RGB -
Newegg links:
CPU: Ryzen 7 7700X -
Cooler: Noctua NH-D15 -
Motherboard: Gigabyte X670 AORUS Elite AX -
RAM: Corsair Vengeance DDR5 64GB 5200mhz -
GPU: XFX Speedster MERC 319 RX 6950 XT -
PSU: MSI MEG Ai1300P PCIE5 1300w -
Case: Fractal Torrent RGB -
Timestamps:
0:00 - Intro
0:51 - How to use Stable Diffusion webui
6:03 - 20 Images in less than a minute
7:14 - Conclusion
#ROCm #amd #stablediffusion #radeon
1 view
37
10
2 weeks ago 00:23:23 1
Create CONSISTENT CHARACTERS from an INPUT IMAGE with FLUX! (ComfyUI Tutorial + Installation Guide)
1 month ago 00:14:26 3
The Survivor
1 month ago 00:04:31 1
Ты меня не любишь, не жалеешь (Сергей Есенин)
1 month ago 00:05:21 1
Лигалайз (.) & Mr. Freeman, “МИР!! ВАШЕМУ!! ДОМУ!!“