3 ms·
> If people could develop AI stuff on consumer cards Technically people can already do that, by leveraging compute shaders in D3D, Vulkan, or WebGPU. I’m certa
by Const-me 3y ago
> If people could develop AI stuff on consumer cards
Technically people can already do that, by leveraging compute shaders in D3D, Vulkan, or WebGPU. I’m certain it’s possible to implement D3D or Vulkan backend for PyTorch, TensorFlow, and similar ML libraries. When I experimented with AI, I did similar stuff myself with C++, C#, and HLSL, and I found it wasn’t terribly hard to do.
However, PyTorch made by Facebook, TensorFlow by Google. It seems most companies who maintain such libraries are only interested in cloud computing. Some of them, like Qualcomm, only care about their own proprietary hardware. None of them seems to care about desktops.