The incumbent approaches usually benefit from a ton of research that might or might not be transferrable to the newcomer. Even if many optimizations also apply

The incumbent approaches usually benefit from a ton of research that might or might not be transferrable to the newcomer. Even if many optimizations also apply to the new approaches, taking advantage of them takes a lot of work. For example, I have not yet implemented KV caches for my nanoGPTs that I'm fooling around with.