[nraynaud]

Naive question: ML tensor libraries don’t use a Z-order memory layout like textures do? It’s not beneficial like it is for textures?

[matusp]

I think that z-order is used to increase speed of loading texture from RAM. But this is not an issue in ML. You usually have all your model weights directly loaded into your GPU memory and you do not need caching for your inputs. At the same time, the entire stack for ML is heavily optimized for other memory layouts already.