When might we see a path from stable diffusion into Get3d (or similar) possible to run on consumer grade hardware? Or are we going to be in ‘8 A100 class GPUs’ for the forseeable future? If the latter, will nvidia or a partner be hosting a cloud-based service to make this tech available to businesses not outfitted with that level of hardware?
Hi, thanks for the questions! We can run inference for GET3D on consumer-grade GPUs if you’re following the instructions from here: GitHub - nv-tlabs/GET3D, which means you can run GET3D on any cloud service following our licence GET3D/LICENSE.txt at master · nv-tlabs/GET3D · GitHub
1 Like
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.