Could you stick a torch.compile in the inference and training code, maybe gated behind a flag? This should help AMD/Nvidia performance (and probably other vendo

Could you stick a torch.compile in the inference and training code, maybe gated behind a flag? This should help AMD/Nvidia performance (and probably other vendors soon) significantly. PyTorch themselves used nanoGPT training as demo for this: https://pytorch.org/blog/accelerating-large-language-models/ https://pytorch.org/blog/accelerating-large-language-models/