The patch is for ggml, which is a generic computing library for AI, not specific to R1. Moreover, the patch is in WASM, which means it is for cases where ggml is compiled to javascript and runs in the browser. In addition, it is an optimization to the CPU back-end and not to the GPU back-end.
Models like R1 are WAY to big to run on CPU. While this patch is amazing, it will not benefit anyone...